{"arm": "dim_no_prov", "item": "CURR-07", "family": "CURR", "rep": 0, "latency_s": 7.93, "context_chars": 4903, "prompt_chars": 5819, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": \"F-064\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Drug allergy valid since 2015, no expiration\"}\n```\n", "obj": {"answer": "penicillin", "source_id": "F-064", "policy_id": null, "unanswerable": false, "stale": false, "note": "Drug allergy valid since 2015, no expiration"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "GAP-02", "family": "GAP", "rep": 0, "latency_s": 7.96, "context_chars": 2097, "prompt_chars": 3010, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain no information about 2016 residence.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain no information about 2016 residence."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "HIST-07", "family": "HIST", "rep": 0, "latency_s": 8.88, "context_chars": 2097, "prompt_chars": 3052, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No gym payment information in records\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No gym payment information in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HIST-04", "family": "HIST", "rep": 0, "latency_s": 9.51, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\"answer\": \"Halvard Systems\", \"source_id\": \"S-EML-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employment period 2018-09-01 to 2021-11-30 covers 2019-10-10.\"}\n```\n", "obj": {"answer": "Halvard Systems", "source_id": "S-EML-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employment period 2018-09-01 to 2021-11-30 covers 2019-10-10."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-05", "family": "CURR", "rep": 0, "latency_s": 9.51, "context_chars": 4903, "prompt_chars": 5813, "raw": "```json\n{\n  \"answer\": \"Aurora 14\",\n  \"source_id\": \"F-070\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Device record valid from 2024-02-20 with open end date.\"\n}\n```\n", "obj": {"answer": "Aurora 14", "source_id": "F-070", "policy_id": null, "unanswerable": false, "stale": false, "note": "Device record valid from 2024-02-20 with open end date."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "GAP-02", "family": "GAP", "rep": 0, "latency_s": 9.61, "context_chars": 4664, "prompt_chars": 5577, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"No residence records available for 2016. Earliest record is 2019.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No residence records available for 2016. Earliest record is 2019."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CONF-03", "family": "CONF", "rep": 0, "latency_s": 13.72, "context_chars": 6744, "prompt_chars": 7688, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records show subscription_cloudline values (9, 12) but lack currency and frequency specification.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show subscription_cloudline values (9, 12) but lack currency and frequency specification."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-08", "family": "HIST", "rep": 1, "latency_s": 15.3, "context_chars": 5242, "prompt_chars": 6195, "raw": "```json\n{\"answer\": \"19\", \"source_id\": \"S-EML-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Trainline pass cost EUR 19/month from 2023-01-01 to 2024-12-31\"}\n```\n", "obj": {"answer": "19", "source_id": "S-EML-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Trainline pass cost EUR 19/month from 2023-01-01 to 2024-12-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HIST-06", "family": "HIST", "rep": 1, "latency_s": 8.6, "context_chars": 4664, "prompt_chars": 5583, "raw": "```json\n{\n  \"answer\": \"vegetarian\",\n  \"source_id\": \"S-SLF-03\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vegetarian diet from 2023-06-02 to 2025-03-29\"\n}\n```\n", "obj": {"answer": "vegetarian", "source_id": "S-SLF-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Vegetarian diet from 2023-06-02 to 2025-03-29"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HOP-06", "family": "HOP", "rep": 1, "latency_s": 9.71, "context_chars": 4903, "prompt_chars": 5852, "raw": "```json\n{\"answer\": \"Tessera Analytics\", \"source_id\": \"F-011\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed 2021-12-01 to 2025-02-28, covering 2024-02-20 purchase date\"}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "F-011", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed 2021-12-01 to 2025-02-28, covering 2024-02-20 purchase date"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-02", "family": "CURR", "rep": 1, "latency_s": 9.34, "context_chars": 10038, "prompt_chars": 10950, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": \"S-SLF-08\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2025-03-30 onward with no end date.\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": "S-SLF-08", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2025-03-30 onward with no end date."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "GAP-03", "family": "GAP", "rep": 0, "latency_s": 7.96, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Vienna trip (2024-09-17) recorded but no cost data in records\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Vienna trip (2024-09-17) recorded but no cost data in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "STALE-02", "family": "STALE", "rep": 1, "latency_s": 14.21, "context_chars": 2097, "prompt_chars": 3020, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": true,\n  \"note\": \"Review was due 2026-08-31; current validity status not recorded.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": true, "note": "Review was due 2026-08-31; current validity status not recorded."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "CURR-03", "family": "CURR", "rep": 1, "latency_s": 7.21, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid until 2027-01-31\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CURR-05", "family": "CURR", "rep": 1, "latency_s": 13.56, "context_chars": 6744, "prompt_chars": 7654, "raw": "```json\n{\"answer\": \"Aurora 14\", \"source_id\": \"S-EML-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Email from Feb 2024; no subsequent updates in records\"}\n```\n", "obj": {"answer": "Aurora 14", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Email from Feb 2024; no subsequent updates in records"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HOP-02", "family": "HOP", "rep": 1, "latency_s": 15.13, "context_chars": 4903, "prompt_chars": 5854, "raw": "```json\n{\"answer\": \"Limassol\", \"source_id\": \"F-001\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"At 12 Harbour Lane, Limassol during Tessera Analytics employment start (2021-12-01).\"}\n```\n", "obj": {"answer": "Limassol", "source_id": "F-001", "policy_id": null, "unanswerable": false, "stale": false, "note": "At 12 Harbour Lane, Limassol during Tessera Analytics employment start (2021-12-01)."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-02", "family": "CURR", "rep": 1, "latency_s": 7.27, "context_chars": 4903, "prompt_chars": 5815, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": \"F-051\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2025-03-30 with no end date.\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": "F-051", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2025-03-30 with no end date."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HOP-02", "family": "HOP", "rep": 1, "latency_s": 10.03, "context_chars": 4664, "prompt_chars": 5615, "raw": "```json\n{\n  \"answer\": \"Limassol\",\n  \"source_id\": \"S-DOC-01\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Started at Tessera Analytics on 2021-12-01, living in Limassol until 2022-07-31.\"\n}\n```\n", "obj": {"answer": "Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Started at Tessera Analytics on 2021-12-01, living in Limassol until 2022-07-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CONF-04", "family": "CONF", "rep": 1, "latency_s": 11.61, "context_chars": 10038, "prompt_chars": 10959, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"S-DEV-01\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Two mobile numbers valid today; device_sync (tier 2) takes precedence over self_reported (tier 4).\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Two mobile numbers valid today; device_sync (tier 2) takes precedence over self_reported (tier 4)."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "STALE-05", "family": "STALE", "rep": 0, "latency_s": 8.05, "context_chars": 5242, "prompt_chars": 6170, "raw": "```json\n{\"answer\": \"Yes, your Aurora 14 laptop is still under warranty until 2027-02-20.\", \"source_id\": \"S-EML-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Warranty expires 2027-02-20, approximately 5 months from today.\"}\n```\n", "obj": {"answer": "Yes, your Aurora 14 laptop is still under warranty until 2027-02-20.", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty expires 2027-02-20, approximately 5 months from today."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "CURR-03", "family": "CURR", "rep": 1, "latency_s": 6.77, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": \"F-061\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid until 2027-01-31\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": "F-061", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CONF-05", "family": "CONF", "rep": 1, "latency_s": 12.14, "context_chars": 6744, "prompt_chars": 7654, "raw": "```json\n{\"answer\": \"0+\", \"source_id\": \"S-DOC-10\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Document scan (tier 1) conflicts with self-reported (tier 4); tier 1 prevails.\"}\n```\n", "obj": {"answer": "0+", "source_id": "S-DOC-10", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Document scan (tier 1) conflicts with self-reported (tier 4); tier 1 prevails."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "GAP-04", "family": "GAP", "rep": 1, "latency_s": 11.85, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister Ilona Vantaa identified, but email address not recorded. Only phone available.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister Ilona Vantaa identified, but email address not recorded. Only phone available."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "ENT-02", "family": "ENT", "rep": 1, "latency_s": 10.73, "context_chars": 4664, "prompt_chars": 5595, "raw": "```json\n{\"answer\": \"Alex Doran is your GP in Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran listed as GP in Larnaca in contacts\"}\n```\n", "obj": {"answer": "Alex Doran is your GP in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran listed as GP in Larnaca in contacts"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "GAP-02", "family": "GAP", "rep": 0, "latency_s": 8.41, "context_chars": 5242, "prompt_chars": 6155, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No residential information available for 2016 in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No residential information available for 2016 in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "HOP-02", "family": "HOP", "rep": 1, "latency_s": 12.0, "context_chars": 10038, "prompt_chars": 10989, "raw": "```json\n{\"answer\": \"Limassol\", \"source_id\": \"S-DOC-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Tessera Analytics employment started 2021-12-01; address valid then was 12 Harbour Lane, Limassol\"}\n```\n", "obj": {"answer": "Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Tessera Analytics employment started 2021-12-01; address valid then was 12 Harbour Lane, Limassol"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HIST-08", "family": "HIST", "rep": 0, "latency_s": 11.98, "context_chars": 10038, "prompt_chars": 10991, "raw": "```json\n{\"answer\": \"19\", \"source_id\": \"S-EML-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Trainline subscription was 19 EUR/month during 2023-07-01.\"}\n```\n", "obj": {"answer": "19", "source_id": "S-EML-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Trainline subscription was 19 EUR/month during 2023-07-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-02", "family": "CURR", "rep": 0, "latency_s": 7.48, "context_chars": 4903, "prompt_chars": 5815, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": \"F-051\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2025-03-30 with no end date\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": "F-051", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2025-03-30 with no end date"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "ENT-01", "family": "ENT", "rep": 1, "latency_s": 9.67, "context_chars": 4903, "prompt_chars": 5847, "raw": "```json\n{\"answer\": \"+357 99 018 442\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Kovacs is the landlord of 7 Kalymnou.\"}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs is the landlord of 7 Kalymnou."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "GAP-01", "family": "GAP", "rep": 1, "latency_s": 7.25, "context_chars": 2097, "prompt_chars": 3022, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"National insurance number not in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "National insurance number not in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CURR-08", "family": "CURR", "rep": 1, "latency_s": 10.29, "context_chars": 6744, "prompt_chars": 7691, "raw": "```json\n{\"answer\": \"11\", \"source_id\": \"S-SLF-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"Record from 2020; current pricing may differ\"}\n```\n", "obj": {"answer": "11", "source_id": "S-SLF-02", "policy_id": null, "unanswerable": false, "stale": true, "note": "Record from 2020; current pricing may differ"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 1, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HIST-04", "family": "HIST", "rep": 1, "latency_s": 9.84, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records lack dated employment history for 2019-10-10.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records lack dated employment history for 2019-10-10."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "CONF-03", "family": "CONF", "rep": 0, "latency_s": 29.65, "context_chars": 4903, "prompt_chars": 5847, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": \"F-080; F-081\",\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Two conflicting Cloudline subscription values: 9 and 12\"\n}\n```\n", "obj": {"answer": null, "source_id": "F-080; F-081", "policy_id": null, "unanswerable": true, "stale": false, "note": "Two conflicting Cloudline subscription values: 9 and 12"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "CURR-01", "family": "CURR", "rep": 1, "latency_s": 10.97, "context_chars": 4664, "prompt_chars": 5579, "raw": "```json\n{\"answer\": \"Kestrel Data Works\", \"source_id\": \"S-DOC-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Joined 2025-03-01, record states still working there.\"}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Joined 2025-03-01, record states still working there."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CONF-05", "family": "CONF", "rep": 0, "latency_s": 30.66, "context_chars": 4903, "prompt_chars": 5813, "raw": "```json\n{\"answer\": \"A+\", \"source_id\": \"F-040\", \"policy_id\": \"prefer-earlier-record\", \"unanswerable\": false, \"stale\": false, \"note\": \"Records F-040 and F-041 conflict (A+ vs 0+). Both valid today.\"}\n```\n", "obj": {"answer": "A+", "source_id": "F-040", "policy_id": "prefer-earlier-record", "unanswerable": false, "stale": false, "note": "Records F-040 and F-041 conflict (A+ vs 0+). Both valid today."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-04", "family": "CURR", "rep": 0, "latency_s": 9.69, "context_chars": 10038, "prompt_chars": 10982, "raw": "```json\n{\n  \"answer\": \"22\",\n  \"source_id\": \"F-084\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Journal subscription 22 EUR/month, active since 2024-03-15\"\n}\n```\n", "obj": {"answer": "22", "source_id": "F-084", "policy_id": null, "unanswerable": false, "stale": false, "note": "Journal subscription 22 EUR/month, active since 2024-03-15"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "ENT-04", "family": "ENT", "rep": 0, "latency_s": 12.27, "context_chars": 6744, "prompt_chars": 7679, "raw": "```json\n{\"answer\": \"3\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, Alex Doran, Alex Kovacs in contacts list\"}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, Alex Kovacs in contacts list"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HIST-05", "family": "HIST", "rep": 0, "latency_s": 10.79, "context_chars": 4664, "prompt_chars": 5592, "raw": "```json\n{\"answer\": \"+357 99 100 224\", \"source_id\": \"S-SLF-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Number held from 2019-01-10 until 2024-06-30\"}\n```\n", "obj": {"answer": "+357 99 100 224", "source_id": "S-SLF-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Number held from 2019-01-10 until 2024-06-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "STALE-02", "family": "STALE", "rep": 1, "latency_s": 14.24, "context_chars": 5242, "prompt_chars": 6165, "raw": "```json\n{\n  \"answer\": \"No, it expired on 2026-08-31.\",\n  \"source_id\": \"S-DOC-08\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Health insurance policy HI-88231 covered 2025-09-01 to 2026-08-31.\"\n}\n```\n", "obj": {"answer": "No, it expired on 2026-08-31.", "source_id": "S-DOC-08", "policy_id": null, "unanswerable": false, "stale": true, "note": "Health insurance policy HI-88231 covered 2025-09-01 to 2026-08-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes", "item": "STALE-03", "family": "STALE", "rep": 1, "latency_s": 7.02, "context_chars": 4664, "prompt_chars": 5593, "raw": "```json\n{\"answer\": \"No\", \"source_id\": \"S-EML-03\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"Corvus 8 warranty ended 2026-07-01\"}\n```\n", "obj": {"answer": "No", "source_id": "S-EML-03", "policy_id": null, "unanswerable": false, "stale": true, "note": "Corvus 8 warranty ended 2026-07-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "flat", "item": "CURR-07", "family": "CURR", "rep": 0, "latency_s": 9.61, "context_chars": 2097, "prompt_chars": 3013, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Drug allergy recorded in personal medical facts\"}\n```\n", "obj": {"answer": "penicillin", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Drug allergy recorded in personal medical facts"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "ENT-02", "family": "ENT", "rep": 0, "latency_s": 10.65, "context_chars": 6744, "prompt_chars": 7675, "raw": "```json\n{\"answer\": \"Alex Doran, in Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran is your general practitioner.\"}\n```\n", "obj": {"answer": "Alex Doran, in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran is your general practitioner."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-02", "family": "CURR", "rep": 0, "latency_s": 10.04, "context_chars": 10038, "prompt_chars": 10950, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": \"F-051\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2025-03-30 onward with no end date\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": "F-051", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2025-03-30 onward with no end date"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-06", "family": "CURR", "rep": 0, "latency_s": 8.47, "context_chars": 10038, "prompt_chars": 10947, "raw": "```json\n{\n  \"answer\": \"CY17 0020 0128 0000 0012 0052 7600\",\n  \"source_id\": \"S-DOC-04\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Document scan captured 2021-11-18, valid from 2021-12-01 onward\"\n}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Document scan captured 2021-11-18, valid from 2021-12-01 onward"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-07", "family": "CURR", "rep": 1, "latency_s": 9.63, "context_chars": 4664, "prompt_chars": 5580, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Penicillin allergy documented 2024-11-09, known since 2015-01-01\"}\n```\n", "obj": {"answer": "penicillin", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Penicillin allergy documented 2024-11-09, known since 2015-01-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "GAP-05", "family": "GAP", "rep": 0, "latency_s": 10.15, "context_chars": 10038, "prompt_chars": 10977, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain employer names only, not office addresses.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain employer names only, not office addresses."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HOP-01", "family": "HOP", "rep": 0, "latency_s": 11.09, "context_chars": 4664, "prompt_chars": 5622, "raw": "```json\n{\"answer\": \"Nadja Lind\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Most recent Nicosia travel was 2026-01-22 with Nadja Lind\"}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent Nicosia travel was 2026-01-22 with Nadja Lind"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "ENT-05", "family": "ENT", "rep": 1, "latency_s": 15.03, "context_chars": 6744, "prompt_chars": 7691, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": \"S-DOC-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner is a colleague at your current employer, Kestrel Data Works.\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner is a colleague at your current employer, Kestrel Data Works."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CONF-04", "family": "CONF", "rep": 0, "latency_s": 17.66, "context_chars": 6744, "prompt_chars": 7665, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": \"F-021\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Two numbers conflicted; lower tier (device_sync) applied.\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "F-021", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Two numbers conflicted; lower tier (device_sync) applied."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "GAP-05", "family": "GAP", "rep": 1, "latency_s": 9.41, "context_chars": 6744, "prompt_chars": 7683, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Employer identified as Kestrel Data Works, but office street address not recorded.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Employer identified as Kestrel Data Works, but office street address not recorded."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "ENT-05", "family": "ENT", "rep": 0, "latency_s": 10.0, "context_chars": 4903, "prompt_chars": 5850, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": \"F-012\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner is listed as colleague at Kestrel Data Works, your current employer since 2025-03-01.\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": "F-012", "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner is listed as colleague at Kestrel Data Works, your current employer since 2025-03-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CONF-06", "family": "CONF", "rep": 0, "latency_s": 29.01, "context_chars": 6744, "prompt_chars": 7710, "raw": "```json\n{\"answer\": \"S-DOC-09 (document_scan tier 1)\", \"source_id\": \"F-031\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Tier 1 outweighs tier 4 per CP-01 conflict rule\"}\n```\n", "obj": {"answer": "S-DOC-09 (document_scan tier 1)", "source_id": "F-031", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Tier 1 outweighs tier 4 per CP-01 conflict rule"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CONF-05", "family": "CONF", "rep": 0, "latency_s": 9.7, "context_chars": 6744, "prompt_chars": 7654, "raw": "```json\n{\"answer\": \"0+\", \"source_id\": \"S-DOC-10\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Conflicting records resolved: document_scan (tier 1) supersedes self_reported (tier 4).\"}\n```\n", "obj": {"answer": "0+", "source_id": "S-DOC-10", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Conflicting records resolved: document_scan (tier 1) supersedes self_reported (tier 4)."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CURR-07", "family": "CURR", "rep": 0, "latency_s": 8.07, "context_chars": 6744, "prompt_chars": 7660, "raw": "```json\n{\n  \"answer\": \"penicillin\",\n  \"source_id\": \"S-DOC-10\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Documented drug allergy from medical record scan.\"\n}\n```\n", "obj": {"answer": "penicillin", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Documented drug allergy from medical record scan."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HIST-03", "family": "HIST", "rep": 1, "latency_s": 8.38, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": \"Tessera Analytics\", \"source_id\": \"F-011\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed by Tessera Analytics from 2021-12-01 to 2025-02-28.\"}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "F-011", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed by Tessera Analytics from 2021-12-01 to 2025-02-28."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "ENT-05", "family": "ENT", "rep": 0, "latency_s": 13.55, "context_chars": 10038, "prompt_chars": 10985, "raw": "```json\n{\n  \"answer\": \"Alex Reiner\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Reiner is listed as colleague at Kestrel Data Works, your current employer since 2025-03-01.\"\n}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner is listed as colleague at Kestrel Data Works, your current employer since 2025-03-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CONF-02", "family": "CONF", "rep": 0, "latency_s": 40.85, "context_chars": 4903, "prompt_chars": 5833, "raw": "```json\n{\n  \"answer\": \"4200\",\n  \"source_id\": \"F-032\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Two conflicting salary records F-031 (4410) and F-032 (4200) valid from 2026-01-01; selected F-032.\"\n}\n```\n", "obj": {"answer": "4200", "source_id": "F-032", "policy_id": null, "unanswerable": false, "stale": false, "note": "Two conflicting salary records F-031 (4410) and F-032 (4200) valid from 2026-01-01; selected F-032."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CONF-05", "family": "CONF", "rep": 1, "latency_s": 16.89, "context_chars": 4903, "prompt_chars": 5813, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Conflicting records: F-040 (A+) and F-041 (0+), both currently valid\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Conflicting records: F-040 (A+) and F-041 (0+), both currently valid"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "GAP-04", "family": "GAP", "rep": 1, "latency_s": 8.09, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister Ilona Vantaa's contact info lacks email address in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister Ilona Vantaa's contact info lacks email address in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "CONF-06", "family": "CONF", "rep": 0, "latency_s": 15.7, "context_chars": 10038, "prompt_chars": 11004, "raw": "```json\n{\n  \"answer\": \"S-DOC-09 (document_scan, tier 1, 4410 EUR) should be believed over S-SLF-07 (self_reported, tier 4, 4200 EUR)\",\n  \"source_id\": \"S-DOC-09\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Lower tier (document_scan=1) wins over higher tier (self_reported=4) per conflict policy.\"\n}\n```\n", "obj": {"answer": "S-DOC-09 (document_scan, tier 1, 4410 EUR) should be believed over S-SLF-07 (self_reported, tier 4, 4200 EUR)", "source_id": "S-DOC-09", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Lower tier (document_scan=1) wins over higher tier (self_reported=4) per conflict policy."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "ENT-03", "family": "ENT", "rep": 1, "latency_s": 6.85, "context_chars": 2097, "prompt_chars": 3049, "raw": "```json\n{\"answer\": \"+357 96 220 145\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner accompanied on 2026-03-14 Berlin trip\"}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner accompanied on 2026-03-14 Berlin trip"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-03", "family": "CURR", "rep": 1, "latency_s": 7.56, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid until 2027-01-31\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-05", "family": "CURR", "rep": 1, "latency_s": 9.67, "context_chars": 5242, "prompt_chars": 6152, "raw": "```json\n{\"answer\": \"Aurora 14\", \"source_id\": \"S-EML-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Purchased 2024-02-20, warranty valid until 2027-02-20\"}\n```\n", "obj": {"answer": "Aurora 14", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Purchased 2024-02-20, warranty valid until 2027-02-20"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-07", "family": "HIST", "rep": 0, "latency_s": 8.23, "context_chars": 6744, "prompt_chars": 7699, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No gym subscription or payment records found in the database.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No gym subscription or payment records found in the database."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "HOP-04", "family": "HOP", "rep": 1, "latency_s": 24.0, "context_chars": 2097, "prompt_chars": 3076, "raw": "```json\n{\n  \"answer\": \"Alex Reiner, Alex Doran, Alex Kovacs, Petros Sarris\",\n  \"source_id\": \"people\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Previous city Larnaca inferred from landlord entry; four contacts live there.\"\n}\n```\n", "obj": {"answer": "Alex Reiner, Alex Doran, Alex Kovacs, Petros Sarris", "source_id": "people", "policy_id": null, "unanswerable": false, "stale": false, "note": "Previous city Larnaca inferred from landlord entry; four contacts live there."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "GAP-05", "family": "GAP", "rep": 0, "latency_s": 7.93, "context_chars": 2097, "prompt_chars": 3036, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Employer is Kestrel Data Works, but office address not in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Employer is Kestrel Data Works, but office address not in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "STALE-04", "family": "STALE", "rep": 0, "latency_s": 7.39, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\n  \"answer\": \"Yes, your driving licence (DL-772104) is valid until 2027-01-31\",\n  \"source_id\": \"S-DOC-07\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Document scan confirms validity through 2027-01-31.\"\n}\n```\n", "obj": {"answer": "Yes, your driving licence (DL-772104) is valid until 2027-01-31", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Document scan confirms validity through 2027-01-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HOP-06", "family": "HOP", "rep": 1, "latency_s": 12.35, "context_chars": 5242, "prompt_chars": 6191, "raw": "```json\n{\n  \"answer\": \"Tessera Analytics\",\n  \"source_id\": \"S-DOC-04\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Employed at Tessera Analytics from 2021-12-01 to 2025-02-28\"\n}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Tessera Analytics from 2021-12-01 to 2025-02-28"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "STALE-01", "family": "STALE", "rep": 1, "latency_s": 8.79, "context_chars": 5242, "prompt_chars": 6219, "raw": "```json\n{\"answer\": \"Passport P4417832, but it cannot be used—it expired 2026-04-11.\", \"source_id\": \"S-DOC-06\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"Passport expired before today; renewal needed\"}\n```\n", "obj": {"answer": "Passport P4417832, but it cannot be used—it expired 2026-04-11.", "source_id": "S-DOC-06", "policy_id": null, "unanswerable": false, "stale": true, "note": "Passport expired before today; renewal needed"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "STALE-05", "family": "STALE", "rep": 1, "latency_s": 10.23, "context_chars": 6744, "prompt_chars": 7672, "raw": "```json\n{\n  \"answer\": \"Yes, your Aurora 14 laptop is under warranty until 2027-02-20.\",\n  \"source_id\": \"F-071\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Warranty expires 2027-02-20, which is after today.\"\n}\n```\n", "obj": {"answer": "Yes, your Aurora 14 laptop is under warranty until 2027-02-20.", "source_id": "F-071", "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty expires 2027-02-20, which is after today."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HIST-04", "family": "HIST", "rep": 0, "latency_s": 8.68, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": \"Halvard Systems\", \"source_id\": \"F-010\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employment valid 2018-09-01 to 2021-11-30 covers 2019-10-10.\"}\n```\n", "obj": {"answer": "Halvard Systems", "source_id": "F-010", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employment valid 2018-09-01 to 2021-11-30 covers 2019-10-10."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "GAP-03", "family": "GAP", "rep": 1, "latency_s": 8.2, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Vienna trip (2024-09-17) documented but cost data absent from records.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Vienna trip (2024-09-17) documented but cost data absent from records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "STALE-05", "family": "STALE", "rep": 0, "latency_s": 12.19, "context_chars": 6744, "prompt_chars": 7672, "raw": "```json\n{\n  \"answer\": \"Yes, your Aurora 14 laptop is under warranty until 2027-02-20.\",\n  \"source_id\": \"F-071\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Warranty expires in ~5.5 months from today.\"\n}\n```\n", "obj": {"answer": "Yes, your Aurora 14 laptop is under warranty until 2027-02-20.", "source_id": "F-071", "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty expires in ~5.5 months from today."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HIST-01", "family": "HIST", "rep": 0, "latency_s": 8.48, "context_chars": 4903, "prompt_chars": 5822, "raw": "```json\n{\"answer\": \"48 Aristotelous, Nicosia\", \"source_id\": \"F-002\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2022-08-01 to 2025-05-14, covering 2023-01-01.\"}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": "F-002", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2022-08-01 to 2025-05-14, covering 2023-01-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CURR-04", "family": "CURR", "rep": 1, "latency_s": 27.73, "context_chars": 6744, "prompt_chars": 7688, "raw": "```json\n{\"answer\": \"22 euro\", \"source_id\": \"S-EML-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"Email from 2024-03-15, subscription prices may have changed\"}\n```\n", "obj": {"answer": "22 euro", "source_id": "S-EML-07", "policy_id": null, "unanswerable": false, "stale": true, "note": "Email from 2024-03-15, subscription prices may have changed"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 1, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "ENT-02", "family": "ENT", "rep": 0, "latency_s": 8.02, "context_chars": 2097, "prompt_chars": 3028, "raw": "```json\n{\n  \"answer\": \"Alex Doran is your doctor, located in Larnaca.\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Doran listed as general practitioner in Larnaca.\"\n}\n```\n", "obj": {"answer": "Alex Doran is your doctor, located in Larnaca.", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran listed as general practitioner in Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HOP-05", "family": "HOP", "rep": 1, "latency_s": 54.92, "context_chars": 4664, "prompt_chars": 5620, "raw": "```json\n{\n  \"answer\": \"99\",\n  \"source_id\": \"S-SLF-02, S-SLF-05, S-EML-06, S-EML-05, S-EML-07\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Five subscriptions active: music EUR 11, Cloudline EUR 12, gym EUR 35, Trainline EUR 19, journal EUR 22.\"\n}\n```\n", "obj": {"answer": "99", "source_id": "S-SLF-02, S-SLF-05, S-EML-06, S-EML-05, S-EML-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Five subscriptions active: music EUR 11, Cloudline EUR 12, gym EUR 35, Trainline EUR 19, journal EUR 22."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HOP-06", "family": "HOP", "rep": 0, "latency_s": 8.52, "context_chars": 2097, "prompt_chars": 3046, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Laptop purchase date not recorded; cannot determine employer on that day\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Laptop purchase date not recorded; cannot determine employer on that day"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "CONF-01", "family": "CONF", "rep": 0, "latency_s": 7.69, "context_chars": 2097, "prompt_chars": 3017, "raw": "```json\n{\"answer\": \"48 Aristotelous, Nicosia\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Home address from current records; review due 2027-05-15.\"}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Home address from current records; review due 2027-05-15."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-01", "family": "CONF", "rep": 0, "latency_s": 46.35, "context_chars": 4664, "prompt_chars": 5584, "raw": "```json\n{\"answer\": \"7 Kalymnou, Larnaca\", \"source_id\": \"S-DOC-03\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Lease valid until 2027-05-15. Conflicting note from 2025-06-20.\"}\n```\n", "obj": {"answer": "7 Kalymnou, Larnaca", "source_id": "S-DOC-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Lease valid until 2027-05-15. Conflicting note from 2025-06-20."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-04", "family": "CURR", "rep": 1, "latency_s": 11.72, "context_chars": 2097, "prompt_chars": 3041, "raw": "```json\n{\"answer\": \"22 euros\", \"source_id\": \"subscription_journal\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Journal subscription value from records\"}\n```\n", "obj": {"answer": "22 euros", "source_id": "subscription_journal", "policy_id": null, "unanswerable": false, "stale": false, "note": "Journal subscription value from records"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "GAP-01", "family": "GAP", "rep": 1, "latency_s": 8.56, "context_chars": 10038, "prompt_chars": 10963, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"National insurance number not found in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "National insurance number not found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CONF-01", "family": "CONF", "rep": 0, "latency_s": 48.35, "context_chars": 6744, "prompt_chars": 7664, "raw": "```json\n{\n  \"answer\": \"7 Kalymnou, Larnaca\",\n  \"source_id\": \"F-003\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Conflicts with F-004; document wins per policy\"\n}\n```\n", "obj": {"answer": "7 Kalymnou, Larnaca", "source_id": "F-003", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Conflicts with F-004; document wins per policy"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "ENT-05", "family": "ENT", "rep": 1, "latency_s": 14.99, "context_chars": 4664, "prompt_chars": 5611, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner is a colleague at Kestrel Data Works, your current employer.\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner is a colleague at Kestrel Data Works, your current employer."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "ENT-05", "family": "ENT", "rep": 0, "latency_s": 12.0, "context_chars": 5242, "prompt_chars": 6189, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner is colleague at Kestrel Data Works in Larnaca\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner is colleague at Kestrel Data Works in Larnaca"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HIST-05", "family": "HIST", "rep": 1, "latency_s": 8.56, "context_chars": 4664, "prompt_chars": 5592, "raw": "```json\n{\"answer\": \"+357 99 100 224\", \"source_id\": \"S-SLF-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Number held from 2019-01-10 until 2024-06-30.\"}\n```\n", "obj": {"answer": "+357 99 100 224", "source_id": "S-SLF-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Number held from 2019-01-10 until 2024-06-30."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CONF-06", "family": "CONF", "rep": 0, "latency_s": 17.58, "context_chars": 5242, "prompt_chars": 6208, "raw": "```json\n{\n  \"answer\": \"The payslip (S-DOC-09) showing EUR 4410 from 2026-01-01 should be believed. Official employment documents are more reliable than personal notes.\",\n  \"source_id\": \"S-DOC-09\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Official payslip record more authoritative than personal note.\"\n}\n```\n", "obj": {"answer": "The payslip (S-DOC-09) showing EUR 4410 from 2026-01-01 should be believed. Official employment documents are more reliable than personal notes.", "source_id": "S-DOC-09", "policy_id": null, "unanswerable": false, "stale": false, "note": "Official payslip record more authoritative than personal note."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "STALE-03", "family": "STALE", "rep": 0, "latency_s": 11.9, "context_chars": 5242, "prompt_chars": 6171, "raw": "```json\n{\"answer\": \"No\", \"source_id\": \"S-EML-03\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Corvus 8 warranty expired 2026-07-01; today is 2026-09-06\"}\n```\n", "obj": {"answer": "No", "source_id": "S-EML-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Corvus 8 warranty expired 2026-07-01; today is 2026-09-06"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dimension", "item": "STALE-05", "family": "STALE", "rep": 1, "latency_s": 8.68, "context_chars": 10038, "prompt_chars": 10966, "raw": "```json\n{\n  \"answer\": \"Yes, your Aurora 14 laptop is under warranty until 2027-02-20\",\n  \"source_id\": \"S-EML-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Warranty remains valid for approximately 5 more months\"\n}\n```\n", "obj": {"answer": "Yes, your Aurora 14 laptop is under warranty until 2027-02-20", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty remains valid for approximately 5 more months"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CURR-03", "family": "CURR", "rep": 0, "latency_s": 8.48, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"From 2017 document scan; licence numbers are permanent identifiers.\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "From 2017 document scan; licence numbers are permanent identifiers."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "GAP-01", "family": "GAP", "rep": 1, "latency_s": 8.55, "context_chars": 6744, "prompt_chars": 7669, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No national insurance number in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No national insurance number in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "ENT-03", "family": "ENT", "rep": 0, "latency_s": 9.31, "context_chars": 10038, "prompt_chars": 10990, "raw": "```json\n{\n  \"answer\": \"+357 96 220 145\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Reiner accompanied Mira to Berlin on 2026-03-14\"\n}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner accompanied Mira to Berlin on 2026-03-14"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CONF-01", "family": "CONF", "rep": 0, "latency_s": 89.16, "context_chars": 5242, "prompt_chars": 6162, "raw": "```json\n{\n  \"answer\": \"48 Aristotelous, Nicosia\",\n  \"source_id\": \"S-SLF-04\",\n  \"policy_id\": \"prefer_personal_note\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Conflicts with S-DOC-03 indicating move.\"\n}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": "S-SLF-04", "policy_id": "prefer_personal_note", "unanswerable": false, "stale": false, "note": "Conflicts with S-DOC-03 indicating move."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-06", "family": "CURR", "rep": 0, "latency_s": 7.7, "context_chars": 4903, "prompt_chars": 5812, "raw": "```json\n{\"answer\": \"CY17 0020 0128 0000 0012 0052 7600\", \"source_id\": \"F-063\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid IBAN from 2021-12-01, still current\"}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "F-063", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid IBAN from 2021-12-01, still current"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HOP-03", "family": "HOP", "rep": 1, "latency_s": 9.93, "context_chars": 5242, "prompt_chars": 6189, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"S-DEV-01\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vienna trip was 2024-09-17. New number active from 2024-07-01.\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Vienna trip was 2024-09-17. New number active from 2024-07-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-05", "family": "CURR", "rep": 0, "latency_s": 7.95, "context_chars": 2097, "prompt_chars": 3007, "raw": "```json\n{\"answer\": \"Aurora 14\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Laptop model Aurora 14 with warranty until 2027-02-20\"}\n```\n", "obj": {"answer": "Aurora 14", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Laptop model Aurora 14 with warranty until 2027-02-20"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-04", "family": "CONF", "rep": 1, "latency_s": 11.41, "context_chars": 4664, "prompt_chars": 5585, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": \"S-DEV-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current number from 2024-07-01, synced from handset.\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Current number from 2024-07-01, synced from handset."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-05", "family": "HIST", "rep": 0, "latency_s": 10.05, "context_chars": 6744, "prompt_chars": 7672, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No records document mobile number on 2023-03-01; earliest captures are from 2024.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No records document mobile number on 2023-03-01; earliest captures are from 2024."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HOP-03", "family": "HOP", "rep": 0, "latency_s": 37.36, "context_chars": 4664, "prompt_chars": 5611, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"S-DEV-01\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vienna trip 2024-09-17. Device record shows new number from 2024-07-01.\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Vienna trip 2024-09-17. Device record shows new number from 2024-07-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-01", "family": "CURR", "rep": 1, "latency_s": 13.69, "context_chars": 5242, "prompt_chars": 6157, "raw": "```json\n{\"answer\": \"Kestrel Data Works\", \"source_id\": \"S-DOC-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Joined 2025-03-01, currently employed per 2025-02-21 note.\"}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Joined 2025-03-01, currently employed per 2025-02-21 note."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "ENT-04", "family": "ENT", "rep": 0, "latency_s": 9.86, "context_chars": 4664, "prompt_chars": 5599, "raw": "```json\n{\n  \"answer\": \"3\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Reiner, Alex Doran, and Alex Kovacs are in your contacts.\"\n}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, and Alex Kovacs are in your contacts."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HOP-01", "family": "HOP", "rep": 0, "latency_s": 10.23, "context_chars": 6744, "prompt_chars": 7702, "raw": "```json\n{\"answer\": \"Nadja Lind\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Most recent Nicosia trip 2026-01-22 with Nadja Lind\"}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent Nicosia trip 2026-01-22 with Nadja Lind"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-03", "family": "CONF", "rep": 0, "latency_s": 61.21, "context_chars": 4664, "prompt_chars": 5608, "raw": "```json\n{\n  \"answer\": \"EUR 12\",\n  \"source_id\": \"S-SLF-05\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Most recent: EUR 12 (Sept 2022). Earlier record said EUR 9 (Apr 2021).\"\n}\n```\n", "obj": {"answer": "EUR 12", "source_id": "S-SLF-05", "policy_id": null, "unanswerable": false, "stale": true, "note": "Most recent: EUR 12 (Sept 2022). Earlier record said EUR 9 (Apr 2021)."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 1, "overclaim": 1, "fabricated": 1, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CONF-01", "family": "CONF", "rep": 1, "latency_s": 62.71, "context_chars": 5242, "prompt_chars": 6162, "raw": "```json\n{\n  \"answer\": \"7 Kalymnou, Larnaca\",\n  \"source_id\": \"S-DOC-03\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Moved to 7 Kalymnou, Larnaca on 2025-05-15 per lease.\"\n}\n```\n", "obj": {"answer": "7 Kalymnou, Larnaca", "source_id": "S-DOC-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Moved to 7 Kalymnou, Larnaca on 2025-05-15 per lease."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "STALE-02", "family": "STALE", "rep": 1, "latency_s": 9.96, "context_chars": 4903, "prompt_chars": 5826, "raw": "```json\n{\"answer\": \"No, health insurance HI-88231 expired on 2026-08-31\", \"source_id\": \"F-062\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"Policy valid until 2026-08-31, today is 2026-09-06.\"}\n```\n", "obj": {"answer": "No, health insurance HI-88231 expired on 2026-08-31", "source_id": "F-062", "policy_id": null, "unanswerable": false, "stale": true, "note": "Policy valid until 2026-08-31, today is 2026-09-06."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dimension", "item": "STALE-04", "family": "STALE", "rep": 1, "latency_s": 10.44, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\"answer\": \"Yes, your driving licence (DL-772104) is valid today. Expires 2027-01-31.\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2017-02-03 to 2027-01-31.\"}\n```\n", "obj": {"answer": "Yes, your driving licence (DL-772104) is valid today. Expires 2027-01-31.", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2017-02-03 to 2027-01-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes", "item": "HOP-03", "family": "HOP", "rep": 1, "latency_s": 10.48, "context_chars": 4664, "prompt_chars": 5611, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"S-DEV-01\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vienna trip was 2024-09-17; new number active from 2024-07-01.\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Vienna trip was 2024-09-17; new number active from 2024-07-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-01", "family": "CURR", "rep": 0, "latency_s": 10.38, "context_chars": 10038, "prompt_chars": 10953, "raw": "```json\n{\"answer\": \"Kestrel Data Works\", \"source_id\": \"S-DOC-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current employer since March 2025 with no end date.\"}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Current employer since March 2025 with no end date."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "ENT-03", "family": "ENT", "rep": 1, "latency_s": 11.5, "context_chars": 4664, "prompt_chars": 5616, "raw": "```json\n{\n  \"answer\": \"+357 96 220 145\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Berlin trip 2026-03-14 with Alex Reiner, colleague at Kestrel Data Works.\"\n}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Berlin trip 2026-03-14 with Alex Reiner, colleague at Kestrel Data Works."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "STALE-03", "family": "STALE", "rep": 1, "latency_s": 8.0, "context_chars": 2097, "prompt_chars": 3026, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No warranty expiration date recorded for phone.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No warranty expiration date recorded for phone."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-01", "family": "HIST", "rep": 1, "latency_s": 10.88, "context_chars": 5242, "prompt_chars": 6161, "raw": "```json\n{\"answer\": \"48 Aristotelous, Nicosia\", \"source_id\": \"S-DOC-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Moved there 2022-08-01, stayed until 2025-05-14.\"}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Moved there 2022-08-01, stayed until 2025-05-14."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HOP-04", "family": "HOP", "rep": 1, "latency_s": 144.37, "context_chars": 5242, "prompt_chars": 6221, "raw": "```json\n{\n  \"answer\": \"Ilona Vantaa and Nadja Lind\",\n  \"source_id\": \"S-DOC-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Immediately previous address was 48 Aristotelous, Nicosia (until 2025-05-14).\"\n}\n```\n", "obj": {"answer": "Ilona Vantaa and Nadja Lind", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Immediately previous address was 48 Aristotelous, Nicosia (until 2025-05-14)."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "GAP-01", "family": "GAP", "rep": 1, "latency_s": 7.06, "context_chars": 4664, "prompt_chars": 5589, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"National insurance number not recorded in available documents\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "National insurance number not recorded in available documents"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "HOP-02", "family": "HOP", "rep": 1, "latency_s": 10.24, "context_chars": 2097, "prompt_chars": 3048, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records show Tessera Analytics as former employer via colleague reference, but no location data for that period.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show Tessera Analytics as former employer via colleague reference, but no location data for that period."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "CURR-06", "family": "CURR", "rep": 0, "latency_s": 13.01, "context_chars": 5242, "prompt_chars": 6151, "raw": "```json\n{\"answer\": \"CY17 0020 0128 0000 0012 0052 7600\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Bank IBAN on file since 2021-12-01.\"}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Bank IBAN on file since 2021-12-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-01", "family": "CURR", "rep": 1, "latency_s": 10.58, "context_chars": 10038, "prompt_chars": 10953, "raw": "```json\n{\n  \"answer\": \"Kestrel Data Works\",\n  \"source_id\": \"S-DOC-05\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Valid from 2025-03-01 with no end date recorded.\"\n}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2025-03-01 with no end date recorded."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HIST-08", "family": "HIST", "rep": 0, "latency_s": 7.36, "context_chars": 2097, "prompt_chars": 3050, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No information about Trainline pass found in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No information about Trainline pass found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CURR-05", "family": "CURR", "rep": 0, "latency_s": 10.58, "context_chars": 6744, "prompt_chars": 7654, "raw": "```json\n{\n  \"answer\": \"Aurora 14\",\n  \"source_id\": \"S-EML-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Aurora 14 laptop documented in email from 2024-02-20; warranty valid until 2027-02-20.\"\n}\n```\n", "obj": {"answer": "Aurora 14", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Aurora 14 laptop documented in email from 2024-02-20; warranty valid until 2027-02-20."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HOP-06", "family": "HOP", "rep": 1, "latency_s": 14.1, "context_chars": 10038, "prompt_chars": 10987, "raw": "```json\n{\n  \"answer\": \"Tessera Analytics\",\n  \"source_id\": \"S-DOC-04\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Aurora 14 purchased 2024-02-20; Tessera Analytics employed from 2021-12-01 to 2025-02-28.\"\n}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Aurora 14 purchased 2024-02-20; Tessera Analytics employed from 2021-12-01 to 2025-02-28."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CONF-05", "family": "CONF", "rep": 1, "latency_s": 10.74, "context_chars": 10038, "prompt_chars": 10948, "raw": "```json\n{\"answer\": \"0+\", \"source_id\": \"F-041\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Document scan (tier 1) overrides self-reported (tier 4) per CP-01\"}\n```\n", "obj": {"answer": "0+", "source_id": "F-041", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Document scan (tier 1) overrides self-reported (tier 4) per CP-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HOP-05", "family": "HOP", "rep": 0, "latency_s": 42.33, "context_chars": 10038, "prompt_chars": 10994, "raw": "```json\n{\"answer\": \"96 EUR\", \"source_id\": \"F-080,F-082,F-083,F-084,F-085\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Five subscriptions active on 2024-06-01; cloudline conflict resolved via CP-01\"}\n```\n", "obj": {"answer": "96 EUR", "source_id": "F-080,F-082,F-083,F-084,F-085", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Five subscriptions active on 2024-06-01; cloudline conflict resolved via CP-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CONF-02", "family": "CONF", "rep": 1, "latency_s": 8.21, "context_chars": 2097, "prompt_chars": 3027, "raw": "```json\n{\"answer\": \"4200\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Monthly salary in EUR from personal records\"}\n```\n", "obj": {"answer": "4200", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Monthly salary in EUR from personal records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "ENT-04", "family": "ENT", "rep": 0, "latency_s": 9.98, "context_chars": 5242, "prompt_chars": 6177, "raw": "```json\n{\"answer\": \"3\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, Alex Doran, and Alex Kovacs are in contacts.\"}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, and Alex Kovacs are in contacts."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "ENT-03", "family": "ENT", "rep": 1, "latency_s": 13.21, "context_chars": 4903, "prompt_chars": 5855, "raw": "```json\n{\"answer\": \"+357 96 220 145\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner travelled to Berlin on 2026-03-14\"}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner travelled to Berlin on 2026-03-14"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CURR-08", "family": "CURR", "rep": 0, "latency_s": 18.4, "context_chars": 6744, "prompt_chars": 7691, "raw": "```json\n{\n  \"answer\": \"11\",\n  \"source_id\": \"S-SLF-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Record from 2020, likely outdated. Units not explicitly specified.\"\n}\n```\n", "obj": {"answer": "11", "source_id": "S-SLF-02", "policy_id": null, "unanswerable": false, "stale": true, "note": "Record from 2020, likely outdated. Units not explicitly specified."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 1, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CONF-04", "family": "CONF", "rep": 0, "latency_s": 9.04, "context_chars": 5242, "prompt_chars": 6163, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": \"S-DEV-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current mobile number since 2024-07-01\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Current mobile number since 2024-07-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "GAP-05", "family": "GAP", "rep": 1, "latency_s": 8.5, "context_chars": 5242, "prompt_chars": 6181, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Employer is Kestrel Data Works, but no office address is recorded.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Employer is Kestrel Data Works, but no office address is recorded."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "CURR-07", "family": "CURR", "rep": 1, "latency_s": 7.83, "context_chars": 5242, "prompt_chars": 6158, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Laboratory confirmed allergy known since 2015.\"}\n```\n", "obj": {"answer": "penicillin", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Laboratory confirmed allergy known since 2015."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "GAP-01", "family": "GAP", "rep": 0, "latency_s": 7.74, "context_chars": 5242, "prompt_chars": 6167, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"National insurance number not recorded in available records\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "National insurance number not recorded in available records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-03", "family": "HIST", "rep": 1, "latency_s": 8.97, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\n  \"answer\": \"Tessera Analytics\",\n  \"source_id\": \"S-DOC-04\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Employed at Tessera Analytics from 2021-12-01 to 2025-02-28\"\n}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Tessera Analytics from 2021-12-01 to 2025-02-28"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-03", "family": "CURR", "rep": 0, "latency_s": 6.09, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Driving licence number from personal records.\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Driving licence number from personal records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HIST-08", "family": "HIST", "rep": 0, "latency_s": 8.1, "context_chars": 4903, "prompt_chars": 5856, "raw": "```json\n{\n  \"answer\": \"19\",\n  \"source_id\": \"F-082\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Trainline subscription was 19 EUR/month from 2023-01-01 to 2024-12-31\"\n}\n```\n", "obj": {"answer": "19", "source_id": "F-082", "policy_id": null, "unanswerable": false, "stale": false, "note": "Trainline subscription was 19 EUR/month from 2023-01-01 to 2024-12-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HOP-01", "family": "HOP", "rep": 1, "latency_s": 13.81, "context_chars": 4664, "prompt_chars": 5622, "raw": "```json\n{\n  \"answer\": \"Nadja Lind\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Most recent trip to Nicosia (sister's city) was 2026-01-22.\"\n}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent trip to Nicosia (sister's city) was 2026-01-22."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-01", "family": "CURR", "rep": 0, "latency_s": 10.7, "context_chars": 4664, "prompt_chars": 5579, "raw": "```json\n{\n  \"answer\": \"Kestrel Data Works\",\n  \"source_id\": \"S-DOC-05\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Joined 2025-03-01, still employed there as of last record\"\n}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Joined 2025-03-01, still employed there as of last record"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "GAP-01", "family": "GAP", "rep": 0, "latency_s": 8.36, "context_chars": 6744, "prompt_chars": 7669, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No national insurance number in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No national insurance number in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HIST-01", "family": "HIST", "rep": 0, "latency_s": 9.21, "context_chars": 4664, "prompt_chars": 5583, "raw": "```json\n{\n  \"answer\": \"48 Aristotelous, Nicosia\",\n  \"source_id\": \"S-DOC-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Moved there 2022-08-01, lived until 2025-05-14.\"\n}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Moved there 2022-08-01, lived until 2025-05-14."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HIST-06", "family": "HIST", "rep": 0, "latency_s": 8.36, "context_chars": 5242, "prompt_chars": 6161, "raw": "```json\n{\n  \"answer\": \"vegetarian\",\n  \"source_id\": \"S-SLF-03\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"2024-01-01 falls within the vegetarian period noted 2023-06-02 to 2025-03-29\"\n}\n```\n", "obj": {"answer": "vegetarian", "source_id": "S-SLF-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "2024-01-01 falls within the vegetarian period noted 2023-06-02 to 2025-03-29"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "STALE-03", "family": "STALE", "rep": 0, "latency_s": 12.39, "context_chars": 4664, "prompt_chars": 5593, "raw": "```json\n{\n  \"answer\": \"No\",\n  \"source_id\": \"S-EML-03\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Corvus 8 warranty ended 2026-07-01; today is 2026-09-06\"\n}\n```\n", "obj": {"answer": "No", "source_id": "S-EML-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Corvus 8 warranty ended 2026-07-01; today is 2026-09-06"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "GAP-02", "family": "GAP", "rep": 1, "latency_s": 7.66, "context_chars": 4903, "prompt_chars": 5816, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Home address records begin from 2019; no data for 2016.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Home address records begin from 2019; no data for 2016."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "STALE-05", "family": "STALE", "rep": 0, "latency_s": 10.11, "context_chars": 2097, "prompt_chars": 3025, "raw": "```json\n{\"answer\": \"Yes, your Aurora 14 laptop warranty is valid until 2027-02-20\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Warranty extends 173 days beyond today\"}\n```\n", "obj": {"answer": "Yes, your Aurora 14 laptop warranty is valid until 2027-02-20", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty extends 173 days beyond today"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "STALE-04", "family": "STALE", "rep": 1, "latency_s": 8.9, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\n  \"answer\": \"Yes, your driving licence DL-772104 is valid today. It runs until 2027-01-31.\",\n  \"source_id\": \"S-DOC-07\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Valid until 2027-01-31.\"\n}\n```\n", "obj": {"answer": "Yes, your driving licence DL-772104 is valid today. It runs until 2027-01-31.", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "CONF-03", "family": "CONF", "rep": 1, "latency_s": 44.59, "context_chars": 5242, "prompt_chars": 6186, "raw": "```json\n{\n  \"answer\": \"12\",\n  \"source_id\": \"S-SLF-05\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Current cost per month in euros; verified in Profile summary\"\n}\n```\n", "obj": {"answer": "12", "source_id": "S-SLF-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Current cost per month in euros; verified in Profile summary"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 1, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "STALE-04", "family": "STALE", "rep": 1, "latency_s": 9.94, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain driving licence number only, no expiration date provided.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain driving licence number only, no expiration date provided."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-01", "family": "HIST", "rep": 0, "latency_s": 11.31, "context_chars": 5242, "prompt_chars": 6161, "raw": "```json\n{\n  \"answer\": \"48 Aristotelous, Nicosia\",\n  \"source_id\": \"S-DOC-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Moved there 2022-08-01, lived until 2025-05-14\"\n}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Moved there 2022-08-01, lived until 2025-05-14"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-04", "family": "HIST", "rep": 1, "latency_s": 12.48, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Only employer record (Kestrel Data Works) is from 2025 document, undated to 2019.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Only employer record (Kestrel Data Works) is from 2025 document, undated to 2019."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-07", "family": "HIST", "rep": 0, "latency_s": 9.66, "context_chars": 5242, "prompt_chars": 6197, "raw": "```json\n{\"answer\": \"35\", \"source_id\": \"S-EML-06\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Gym active Sept 2022–May 2025\"}\n```\n", "obj": {"answer": "35", "source_id": "S-EML-06", "policy_id": null, "unanswerable": false, "stale": false, "note": "Gym active Sept 2022–May 2025"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "STALE-02", "family": "STALE", "rep": 0, "latency_s": 32.15, "context_chars": 10038, "prompt_chars": 10961, "raw": "```json\n{\n  \"answer\": \"No, your health insurance is not valid today. Policy HI-88231 expired on 2026-08-31.\",\n  \"source_id\": \"F-062\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Policy expired on 2026-08-31; today is 2026-09-06\"\n}\n```\n", "obj": {"answer": "No, your health insurance is not valid today. Policy HI-88231 expired on 2026-08-31.", "source_id": "F-062", "policy_id": null, "unanswerable": false, "stale": true, "note": "Policy expired on 2026-08-31; today is 2026-09-06"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "STALE-01", "family": "STALE", "rep": 0, "latency_s": 8.49, "context_chars": 4903, "prompt_chars": 5880, "raw": "```json\n{\n  \"answer\": \"Your passport number is P4417832, but it expired on 2026-04-11 and cannot be used.\",\n  \"source_id\": \"F-060\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Passport expired April 2026. Renewal required before travel.\"\n}\n```\n", "obj": {"answer": "Your passport number is P4417832, but it expired on 2026-04-11 and cannot be used.", "source_id": "F-060", "policy_id": null, "unanswerable": false, "stale": true, "note": "Passport expired April 2026. Renewal required before travel."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HOP-03", "family": "HOP", "rep": 0, "latency_s": 44.3, "context_chars": 4903, "prompt_chars": 5850, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"F-021\",\n  \"policy_id\": \"most-recent-change-date\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vienna trip 2024-09-17; F-021 and F-022 both valid that date\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "F-021", "policy_id": "most-recent-change-date", "unanswerable": false, "stale": false, "note": "Vienna trip 2024-09-17; F-021 and F-022 both valid that date"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CONF-03", "family": "CONF", "rep": 0, "latency_s": 18.48, "context_chars": 10038, "prompt_chars": 10982, "raw": "```json\n{\"answer\": \"9\", \"source_id\": \"S-EML-04\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Email source (9) wins over self-reported (12); both valid from 2021-05-01\"}\n```\n", "obj": {"answer": "9", "source_id": "S-EML-04", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Email source (9) wins over self-reported (12); both valid from 2021-05-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-05", "family": "CURR", "rep": 0, "latency_s": 20.54, "context_chars": 5242, "prompt_chars": 6152, "raw": "```json\n{\n  \"answer\": \"Aurora 14\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Purchased 2024-02-20; warranty valid until 2027-02-20.\"\n}\n```\n", "obj": {"answer": "Aurora 14", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Purchased 2024-02-20; warranty valid until 2027-02-20."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-03", "family": "HIST", "rep": 1, "latency_s": 10.31, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Only employer record is from 2025-02-21; no data for 2022-05-01.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Only employer record is from 2025-02-21; no data for 2022-05-01."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-03", "family": "HIST", "rep": 0, "latency_s": 9.45, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": \"Tessera Analytics\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employment from 2021-12-01 to 2025-02-28 confirmed.\"}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employment from 2021-12-01 to 2025-02-28 confirmed."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "ENT-05", "family": "ENT", "rep": 0, "latency_s": 13.14, "context_chars": 4664, "prompt_chars": 5611, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner is a colleague at Kestrel Data Works, Mira's current employer.\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner is a colleague at Kestrel Data Works, Mira's current employer."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "STALE-05", "family": "STALE", "rep": 1, "latency_s": 8.63, "context_chars": 2097, "prompt_chars": 3025, "raw": "```json\n{\"answer\": \"Yes, your Aurora 14 laptop is still under warranty until 2027-02-20.\", \"source_id\": \"laptop_warranty_until\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Warranty expires in approximately 5 months from today.\"}\n```\n", "obj": {"answer": "Yes, your Aurora 14 laptop is still under warranty until 2027-02-20.", "source_id": "laptop_warranty_until", "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty expires in approximately 5 months from today."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dimension", "item": "STALE-03", "family": "STALE", "rep": 0, "latency_s": 9.87, "context_chars": 10038, "prompt_chars": 10967, "raw": "```json\n{\"answer\": \"No, the phone warranty expired on 2026-07-01\", \"source_id\": \"F-073\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"Warranty period ended 67 days ago\"}\n```\n", "obj": {"answer": "No, the phone warranty expired on 2026-07-01", "source_id": "F-073", "policy_id": null, "unanswerable": false, "stale": true, "note": "Warranty period ended 67 days ago"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "CONF-06", "family": "CONF", "rep": 0, "latency_s": 20.09, "context_chars": 4903, "prompt_chars": 5869, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"F-031 (4410 EUR) and F-032 (4200 EUR) both valid 2026-01-01 onwards; no conflict resolution rule\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "F-031 (4410 EUR) and F-032 (4200 EUR) both valid 2026-01-01 onwards; no conflict resolution rule"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HIST-01", "family": "HIST", "rep": 1, "latency_s": 7.42, "context_chars": 6744, "prompt_chars": 7663, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"No records contain home address data from 2023 or earlier.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No records contain home address data from 2023 or earlier."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "GAP-02", "family": "GAP", "rep": 1, "latency_s": 9.4, "context_chars": 6744, "prompt_chars": 7657, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No address records found covering 2016. Earliest home_address fact is from 2025.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No address records found covering 2016. Earliest home_address fact is from 2025."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "GAP-03", "family": "GAP", "rep": 1, "latency_s": 6.45, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Vienna trip documented 2024-09-17 but cost not recorded\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Vienna trip documented 2024-09-17 but cost not recorded"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-04", "family": "HIST", "rep": 0, "latency_s": 8.08, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": \"Halvard Systems\", \"source_id\": \"S-EML-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed there 2018-09-01 to 2021-11-30\"}\n```\n", "obj": {"answer": "Halvard Systems", "source_id": "S-EML-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed there 2018-09-01 to 2021-11-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-08", "family": "CURR", "rep": 0, "latency_s": 14.86, "context_chars": 4664, "prompt_chars": 5611, "raw": "```json\n{\n  \"answer\": \"EUR 11\",\n  \"source_id\": \"S-SLF-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Music service costs EUR 11/month since 2020-01-01, no cancellation recorded\"\n}\n```\n", "obj": {"answer": "EUR 11", "source_id": "S-SLF-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Music service costs EUR 11/month since 2020-01-01, no cancellation recorded"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HOP-04", "family": "HOP", "rep": 1, "latency_s": 15.81, "context_chars": 10038, "prompt_chars": 11017, "raw": "```json\n{\n  \"answer\": \"Ilona Vantaa, Nadja Lind\",\n  \"source_id\": \"F-002\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Both live in Nicosia, Mira's address 2022-08-01 to 2025-05-14.\"\n}\n```\n", "obj": {"answer": "Ilona Vantaa, Nadja Lind", "source_id": "F-002", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Both live in Nicosia, Mira's address 2022-08-01 to 2025-05-14."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "GAP-03", "family": "GAP", "rep": 0, "latency_s": 9.08, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Trip documented on 2024-09-17 but cost not recorded.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Trip documented on 2024-09-17 but cost not recorded."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CONF-02", "family": "CONF", "rep": 1, "latency_s": 31.61, "context_chars": 6744, "prompt_chars": 7674, "raw": "```json\n{\"answer\": \"4410\", \"source_id\": \"F-031\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Document salary from Jan 2026; self-reported (4200) from Feb 2026 conflicts\"}\n```\n", "obj": {"answer": "4410", "source_id": "F-031", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Document salary from Jan 2026; self-reported (4200) from Feb 2026 conflicts"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "STALE-02", "family": "STALE", "rep": 1, "latency_s": 11.26, "context_chars": 10038, "prompt_chars": 10961, "raw": "```json\n{\n  \"answer\": \"No, it expired on 2026-08-31.\",\n  \"source_id\": \"S-DOC-08\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Policy HI-88231 valid until 2026-08-31; today is 2026-09-06.\"\n}\n```\n", "obj": {"answer": "No, it expired on 2026-08-31.", "source_id": "S-DOC-08", "policy_id": null, "unanswerable": false, "stale": true, "note": "Policy HI-88231 valid until 2026-08-31; today is 2026-09-06."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "STALE-05", "family": "STALE", "rep": 1, "latency_s": 8.5, "context_chars": 4903, "prompt_chars": 5831, "raw": "```json\n{\"answer\": \"Yes, still under warranty until 2027-02-20\", \"source_id\": \"F-071\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Aurora 14 warranty expires 2027-02-20\"}\n```\n", "obj": {"answer": "Yes, still under warranty until 2027-02-20", "source_id": "F-071", "policy_id": null, "unanswerable": false, "stale": false, "note": "Aurora 14 warranty expires 2027-02-20"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes", "item": "STALE-05", "family": "STALE", "rep": 1, "latency_s": 8.63, "context_chars": 4664, "prompt_chars": 5592, "raw": "```json\n{\n  \"answer\": \"Yes, your Aurora 14 laptop is still under warranty until 2027-02-20.\",\n  \"source_id\": \"S-EML-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Warranty purchased with laptop on 2024-02-20\"\n}\n```\n", "obj": {"answer": "Yes, your Aurora 14 laptop is still under warranty until 2027-02-20.", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty purchased with laptop on 2024-02-20"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "STALE-01", "family": "STALE", "rep": 0, "latency_s": 8.63, "context_chars": 6744, "prompt_chars": 7721, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Passport number not found in records for Mira Vantaa.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Passport number not found in records for Mira Vantaa."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "STALE-03", "family": "STALE", "rep": 1, "latency_s": 35.59, "context_chars": 10038, "prompt_chars": 10967, "raw": "```json\n{\n  \"answer\": \"No\",\n  \"source_id\": \"F-073\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Warranty expired July 1, 2026; today is September 6, 2026\"\n}\n```\n", "obj": {"answer": "No", "source_id": "F-073", "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty expired July 1, 2026; today is September 6, 2026"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CONF-06", "family": "CONF", "rep": 1, "latency_s": 15.28, "context_chars": 6744, "prompt_chars": 7710, "raw": "```json\n{\n  \"answer\": \"S-DOC-09 (4410 EUR from document scan) should be believed. Document scans (tier 1) outrank self-reported data (tier 4) per the conflict policy.\",\n  \"source_id\": \"S-DOC-09\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Document scan source tier 1 beats self-reported tier 4.\"\n}\n```\n", "obj": {"answer": "S-DOC-09 (4410 EUR from document scan) should be believed. Document scans (tier 1) outrank self-reported data (tier 4) per the conflict policy.", "source_id": "S-DOC-09", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Document scan source tier 1 beats self-reported tier 4."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-06", "family": "CURR", "rep": 1, "latency_s": 7.54, "context_chars": 4903, "prompt_chars": 5812, "raw": "```json\n{\"answer\": \"CY17 0020 0128 0000 0012 0052 7600\", \"source_id\": \"F-063\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid since 2021-12-01, no expiration date\"}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "F-063", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid since 2021-12-01, no expiration date"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-06", "family": "CURR", "rep": 0, "latency_s": 7.96, "context_chars": 2097, "prompt_chars": 3006, "raw": "```json\n{\n  \"answer\": \"CY17 0020 0128 0000 0012 0052 7600\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Bank IBAN from personal records\"\n}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Bank IBAN from personal records"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-08", "family": "CURR", "rep": 0, "latency_s": 20.81, "context_chars": 2097, "prompt_chars": 3044, "raw": "```json\n{\"answer\": \"11 euro\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Music subscription monthly cost from personal records.\"}\n```\n", "obj": {"answer": "11 euro", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Music subscription monthly cost from personal records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-06", "family": "CURR", "rep": 1, "latency_s": 8.25, "context_chars": 5242, "prompt_chars": 6151, "raw": "```json\n{\"answer\": \"CY17 0020 0128 0000 0012 0052 7600\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Bank IBAN on file since 2021-12-01\"}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Bank IBAN on file since 2021-12-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-02", "family": "HIST", "rep": 1, "latency_s": 8.55, "context_chars": 6744, "prompt_chars": 7663, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Home address records exist but lack date-of-residence information for 2020-06-15.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Home address records exist but lack date-of-residence information for 2020-06-15."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "GAP-04", "family": "GAP", "rep": 0, "latency_s": 6.97, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister's email address not recorded in personal records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister's email address not recorded in personal records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-08", "family": "HIST", "rep": 0, "latency_s": 7.57, "context_chars": 5242, "prompt_chars": 6195, "raw": "```json\n{\"answer\": \"19\", \"source_id\": \"S-EML-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Trainline pass cost EUR 19/month from 2023-01-01 to 2024-12-31\"}\n```\n", "obj": {"answer": "19", "source_id": "S-EML-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Trainline pass cost EUR 19/month from 2023-01-01 to 2024-12-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CONF-04", "family": "CONF", "rep": 1, "latency_s": 18.25, "context_chars": 5242, "prompt_chars": 6163, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": \"S-DEV-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current number from July 2024\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Current number from July 2024"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "GAP-02", "family": "GAP", "rep": 1, "latency_s": 8.08, "context_chars": 4664, "prompt_chars": 5577, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No records document residence before 2019.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No records document residence before 2019."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HOP-02", "family": "HOP", "rep": 0, "latency_s": 13.24, "context_chars": 4903, "prompt_chars": 5854, "raw": "```json\n{\n  \"answer\": \"Limassol\",\n  \"source_id\": \"F-001\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Tessera Analytics employment began 2021-12-01; F-001 address valid through 2022-07-31.\"\n}\n```\n", "obj": {"answer": "Limassol", "source_id": "F-001", "policy_id": null, "unanswerable": false, "stale": false, "note": "Tessera Analytics employment began 2021-12-01; F-001 address valid through 2022-07-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "GAP-03", "family": "GAP", "rep": 1, "latency_s": 8.91, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Vienna trip recorded (2024-09-17) but cost not in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Vienna trip recorded (2024-09-17) but cost not in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "GAP-03", "family": "GAP", "rep": 0, "latency_s": 7.15, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Vienna trip mentioned 2024-09-17 but cost not recorded\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Vienna trip mentioned 2024-09-17 but cost not recorded"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "HIST-03", "family": "HIST", "rep": 0, "latency_s": 8.49, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records show current employer but lack dated employment history.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show current employer but lack dated employment history."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "HIST-02", "family": "HIST", "rep": 1, "latency_s": 7.07, "context_chars": 2097, "prompt_chars": 3016, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain only current address; no historical address data for 2020.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain only current address; no historical address data for 2020."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "CONF-01", "family": "CONF", "rep": 1, "latency_s": 17.99, "context_chars": 4664, "prompt_chars": 5584, "raw": "```json\n{\"answer\": \"7 Kalymnou, Larnaca\", \"source_id\": \"S-DOC-03\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Official lease takes precedence over contradicting personal note S-SLF-04.\"}\n```\n", "obj": {"answer": "7 Kalymnou, Larnaca", "source_id": "S-DOC-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Official lease takes precedence over contradicting personal note S-SLF-04."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HIST-07", "family": "HIST", "rep": 1, "latency_s": 8.11, "context_chars": 4903, "prompt_chars": 5858, "raw": "```json\n{\"answer\": \"35\", \"source_id\": \"F-083\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Gym subscription record valid 2022-09-01 to 2025-05-31\"}\n```\n", "obj": {"answer": "35", "source_id": "F-083", "policy_id": null, "unanswerable": false, "stale": false, "note": "Gym subscription record valid 2022-09-01 to 2025-05-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-07", "family": "HIST", "rep": 1, "latency_s": 8.82, "context_chars": 6744, "prompt_chars": 7699, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"No gym payment data found in records.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No gym payment data found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "CURR-05", "family": "CURR", "rep": 1, "latency_s": 10.51, "context_chars": 10038, "prompt_chars": 10948, "raw": "```json\n{\"answer\": \"Aurora 14\", \"source_id\": \"S-EML-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Laptop model owned since February 2024.\"}\n```\n", "obj": {"answer": "Aurora 14", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Laptop model owned since February 2024."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HOP-05", "family": "HOP", "rep": 1, "latency_s": 51.52, "context_chars": 10038, "prompt_chars": 10994, "raw": "```json\n{\n  \"answer\": \"96 euro\",\n  \"source_id\": \"S-EML-04, S-EML-05, S-EML-06, S-EML-07, S-SLF-02\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Five active subscriptions on 2024-06-01. Cloudline conflict resolved by policy.\"\n}\n```\n", "obj": {"answer": "96 euro", "source_id": "S-EML-04, S-EML-05, S-EML-06, S-EML-07, S-SLF-02", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Five active subscriptions on 2024-06-01. Cloudline conflict resolved by policy."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HOP-01", "family": "HOP", "rep": 0, "latency_s": 12.17, "context_chars": 5242, "prompt_chars": 6200, "raw": "```json\n{\"answer\": \"Nadja Lind\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Most recent travel to Nicosia was 2026-01-22.\"}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent travel to Nicosia was 2026-01-22."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CURR-04", "family": "CURR", "rep": 0, "latency_s": 11.19, "context_chars": 6744, "prompt_chars": 7688, "raw": "```json\n{\"answer\": \"22\", \"source_id\": \"F-084\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"Record from 2024-03-15; current pricing may differ.\"}\n```\n", "obj": {"answer": "22", "source_id": "F-084", "policy_id": null, "unanswerable": false, "stale": true, "note": "Record from 2024-03-15; current pricing may differ."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 1, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "GAP-03", "family": "GAP", "rep": 1, "latency_s": 7.34, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Records show Vienna trip on 2024-09-17 but no cost information\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show Vienna trip on 2024-09-17 but no cost information"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CURR-01", "family": "CURR", "rep": 0, "latency_s": 18.44, "context_chars": 6744, "prompt_chars": 7659, "raw": "```json\n{\"answer\": \"Kestrel Data Works\", \"source_id\": \"S-DOC-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Document scan from 2025-02-21; no recent update available.\"}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Document scan from 2025-02-21; no recent update available."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HIST-05", "family": "HIST", "rep": 1, "latency_s": 9.23, "context_chars": 4903, "prompt_chars": 5831, "raw": "```json\n{\"answer\": \"+357 99 100 224\", \"source_id\": \"F-020\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Mobile number valid from 2019-01-10 to 2024-06-30.\"}\n```\n", "obj": {"answer": "+357 99 100 224", "source_id": "F-020", "policy_id": null, "unanswerable": false, "stale": false, "note": "Mobile number valid from 2019-01-10 to 2024-06-30."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HOP-06", "family": "HOP", "rep": 0, "latency_s": 9.13, "context_chars": 4903, "prompt_chars": 5852, "raw": "```json\n{\"answer\": \"Tessera Analytics\", \"source_id\": \"F-011\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Aurora 14 laptop purchased 2024-02-20; employed at Tessera Analytics then.\"}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "F-011", "policy_id": null, "unanswerable": false, "stale": false, "note": "Aurora 14 laptop purchased 2024-02-20; employed at Tessera Analytics then."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HIST-04", "family": "HIST", "rep": 1, "latency_s": 10.37, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": \"Halvard Systems\", \"source_id\": \"S-EML-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed at Halvard Systems from 2018-09-01 to 2021-11-30\"}\n```\n", "obj": {"answer": "Halvard Systems", "source_id": "S-EML-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Halvard Systems from 2018-09-01 to 2021-11-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HOP-03", "family": "HOP", "rep": 0, "latency_s": 11.55, "context_chars": 10038, "prompt_chars": 10985, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"S-DEV-01\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vienna trip was 2024-09-17. Device sync (tier 2) outranks self-reported (tier 4).\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Vienna trip was 2024-09-17. Device sync (tier 2) outranks self-reported (tier 4)."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-02", "family": "CURR", "rep": 1, "latency_s": 9.28, "context_chars": 5242, "prompt_chars": 6154, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": \"S-SLF-08\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Pescatarian since 2025-03-30; vegetarian before that from 2023-06-02\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": "S-SLF-08", "policy_id": null, "unanswerable": false, "stale": false, "note": "Pescatarian since 2025-03-30; vegetarian before that from 2023-06-02"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HOP-03", "family": "HOP", "rep": 0, "latency_s": 11.82, "context_chars": 6744, "prompt_chars": 7691, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"S-DEV-01\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vienna trip 2024-09-17; lower tier device_sync record wins\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Vienna trip 2024-09-17; lower tier device_sync record wins"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "STALE-04", "family": "STALE", "rep": 0, "latency_s": 7.5, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain licence number only, not expiration date or validity status.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain licence number only, not expiration date or validity status."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HOP-05", "family": "HOP", "rep": 0, "latency_s": 27.43, "context_chars": 5242, "prompt_chars": 6198, "raw": "```json\n{\"answer\": \"EUR 99\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Five active subscriptions: music, storage, gym, Trainline, journal.\"}\n```\n", "obj": {"answer": "EUR 99", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Five active subscriptions: music, storage, gym, Trainline, journal."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "STALE-04", "family": "STALE", "rep": 1, "latency_s": 7.73, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": \"Yes, your driving licence DL-772104 is valid until 2027-01-31.\", \"source_id\": \"F-061\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid until 2027-01-31.\"}\n```\n", "obj": {"answer": "Yes, your driving licence DL-772104 is valid until 2027-01-31.", "source_id": "F-061", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dimension", "item": "GAP-02", "family": "GAP", "rep": 1, "latency_s": 7.9, "context_chars": 10038, "prompt_chars": 10951, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No home address records for 2016. Earliest record is 2019-03-01.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No home address records for 2016. Earliest record is 2019-03-01."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HIST-01", "family": "HIST", "rep": 0, "latency_s": 9.21, "context_chars": 6744, "prompt_chars": 7663, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"No records cover 2023-01-01. Earliest home address records from 2025.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No records cover 2023-01-01. Earliest home address records from 2025."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "GAP-03", "family": "GAP", "rep": 0, "latency_s": 9.49, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Vienna trip (2024-09-17) is recorded but cost information is not in the records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Vienna trip (2024-09-17) is recorded but cost information is not in the records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "STALE-05", "family": "STALE", "rep": 0, "latency_s": 7.49, "context_chars": 4903, "prompt_chars": 5831, "raw": "```json\n{\"answer\": \"Yes, your Aurora 14 laptop is still under warranty until 2027-02-20.\", \"source_id\": \"F-071\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Warranty expires 2027-02-20.\"}\n```\n", "obj": {"answer": "Yes, your Aurora 14 laptop is still under warranty until 2027-02-20.", "source_id": "F-071", "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty expires 2027-02-20."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HIST-05", "family": "HIST", "rep": 1, "latency_s": 8.82, "context_chars": 6744, "prompt_chars": 7672, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No mobile phone record predates 2023-03-01. Earliest capture is 2024-05-19.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No mobile phone record predates 2023-03-01. Earliest capture is 2024-05-19."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "ENT-01", "family": "ENT", "rep": 1, "latency_s": 9.17, "context_chars": 10038, "prompt_chars": 10982, "raw": "```json\n{\n  \"answer\": \"+357 99 018 442\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Kovacs is landlord of 7 Kalymnou, current residence.\"\n}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs is landlord of 7 Kalymnou, current residence."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HIST-05", "family": "HIST", "rep": 1, "latency_s": 8.23, "context_chars": 2097, "prompt_chars": 3025, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain only current mobile number; no historical data for 2023-03-01.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain only current mobile number; no historical data for 2023-03-01."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "HIST-08", "family": "HIST", "rep": 1, "latency_s": 7.42, "context_chars": 2097, "prompt_chars": 3050, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No Trainline pass information in records\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No Trainline pass information in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HIST-03", "family": "HIST", "rep": 0, "latency_s": 9.1, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": \"Tessera Analytics\", \"source_id\": \"F-011\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Tessera Analytics employment period: 2021-12-01 to 2025-02-28.\"}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "F-011", "policy_id": null, "unanswerable": false, "stale": false, "note": "Tessera Analytics employment period: 2021-12-01 to 2025-02-28."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-05", "family": "CURR", "rep": 1, "latency_s": 8.29, "context_chars": 2097, "prompt_chars": 3007, "raw": "```json\n{\"answer\": \"Aurora 14\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Device laptop attribute from personal records.\"}\n```\n", "obj": {"answer": "Aurora 14", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Device laptop attribute from personal records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "GAP-05", "family": "GAP", "rep": 1, "latency_s": 9.32, "context_chars": 10038, "prompt_chars": 10977, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records show current employer is Kestrel Data Works but no office address provided.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show current employer is Kestrel Data Works but no office address provided."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "ENT-03", "family": "ENT", "rep": 1, "latency_s": 9.55, "context_chars": 6744, "prompt_chars": 7696, "raw": "```json\n{\"answer\": \"+357 96 220 145\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner accompanied you to Berlin on 2026-03-14\"}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner accompanied you to Berlin on 2026-03-14"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "ENT-05", "family": "ENT", "rep": 0, "latency_s": 7.44, "context_chars": 2097, "prompt_chars": 3044, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Listed as colleague at Kestrel Data Works\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Listed as colleague at Kestrel Data Works"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-03", "family": "CURR", "rep": 0, "latency_s": 6.92, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\n  \"answer\": \"DL-772104\",\n  \"source_id\": \"F-061\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Valid until 2027-01-31\"\n}\n```\n", "obj": {"answer": "DL-772104", "source_id": "F-061", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-08", "family": "CURR", "rep": 1, "latency_s": 10.95, "context_chars": 2097, "prompt_chars": 3044, "raw": "```json\n{\"answer\": \"11\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Monthly music subscription cost from records.\"}\n```\n", "obj": {"answer": "11", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Monthly music subscription cost from records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HOP-06", "family": "HOP", "rep": 0, "latency_s": 53.75, "context_chars": 6744, "prompt_chars": 7693, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Aurora 14 purchase date not explicitly stated in records\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Aurora 14 purchase date not explicitly stated in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "HOP-01", "family": "HOP", "rep": 1, "latency_s": 8.42, "context_chars": 2097, "prompt_chars": 3055, "raw": "```json\n{\"answer\": \"Nadja Lind\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Most recent trip to Nicosia was 2026-01-22 with Nadja Lind\"}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent trip to Nicosia was 2026-01-22 with Nadja Lind"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "ENT-03", "family": "ENT", "rep": 0, "latency_s": 9.45, "context_chars": 6744, "prompt_chars": 7696, "raw": "```json\n{\"answer\": \"+357 96 220 145\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner traveled to Berlin on 2026-03-14.\"}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner traveled to Berlin on 2026-03-14."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-03", "family": "CURR", "rep": 0, "latency_s": 9.36, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid until 2027-01-31\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HIST-02", "family": "HIST", "rep": 0, "latency_s": 7.22, "context_chars": 2097, "prompt_chars": 3016, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain only current address, no historical data for 2020.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain only current address, no historical data for 2020."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "ENT-04", "family": "ENT", "rep": 1, "latency_s": 9.23, "context_chars": 10038, "prompt_chars": 10973, "raw": "```json\n{\"answer\": \"3\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, Alex Doran, Alex Kovacs in contacts.\"}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, Alex Kovacs in contacts."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HOP-05", "family": "HOP", "rep": 0, "latency_s": 42.04, "context_chars": 6744, "prompt_chars": 7700, "raw": "```json\n{\n  \"answer\": \"45 EUR\",\n  \"source_id\": \"F-081, F-084, F-085\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Cloudline 12 + journal 22 + music 11 EUR from most recent captures\"\n}\n```\n", "obj": {"answer": "45 EUR", "source_id": "F-081, F-084, F-085", "policy_id": null, "unanswerable": false, "stale": false, "note": "Cloudline 12 + journal 22 + music 11 EUR from most recent captures"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HIST-08", "family": "HIST", "rep": 0, "latency_s": 10.26, "context_chars": 4664, "prompt_chars": 5617, "raw": "```json\n{\"answer\": \"19\", \"source_id\": \"S-EML-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Trainline pass active 2023-01-01 to 2024-12-31 at EUR 19/month\"}\n```\n", "obj": {"answer": "19", "source_id": "S-EML-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Trainline pass active 2023-01-01 to 2024-12-31 at EUR 19/month"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-02", "family": "CONF", "rep": 0, "latency_s": 16.23, "context_chars": 4664, "prompt_chars": 5594, "raw": "```json\n{\"answer\": \"EUR 4410\", \"source_id\": \"S-DOC-09\", \"policy_id\": \"prefer_official_payslip_over_personal_note\", \"unanswerable\": false, \"stale\": false, \"note\": \"Payslip (S-DOC-09) shows EUR 4410 from 2026-01-01; conflicts with personal note (S-SLF-07).\"}\n```\n", "obj": {"answer": "EUR 4410", "source_id": "S-DOC-09", "policy_id": "prefer_official_payslip_over_personal_note", "unanswerable": false, "stale": false, "note": "Payslip (S-DOC-09) shows EUR 4410 from 2026-01-01; conflicts with personal note (S-SLF-07)."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HIST-05", "family": "HIST", "rep": 0, "latency_s": 7.55, "context_chars": 10038, "prompt_chars": 10966, "raw": "```json\n{\n  \"answer\": \"+357 99 100 224\",\n  \"source_id\": \"S-SLF-01\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Mobile phone valid 2019-01-10 to 2024-06-30\"\n}\n```\n", "obj": {"answer": "+357 99 100 224", "source_id": "S-SLF-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Mobile phone valid 2019-01-10 to 2024-06-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CONF-01", "family": "CONF", "rep": 1, "latency_s": 12.26, "context_chars": 10038, "prompt_chars": 10958, "raw": "```json\n{\n  \"answer\": \"7 Kalymnou, Larnaca\",\n  \"source_id\": \"S-DOC-03\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Document source (tier 1) prevails over conflicting self-report (tier 4).\"\n}\n```\n", "obj": {"answer": "7 Kalymnou, Larnaca", "source_id": "S-DOC-03", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Document source (tier 1) prevails over conflicting self-report (tier 4)."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "GAP-03", "family": "GAP", "rep": 1, "latency_s": 7.39, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records confirm Vienna trip on 2024-09-17 but no cost information provided.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records confirm Vienna trip on 2024-09-17 but no cost information provided."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "GAP-01", "family": "GAP", "rep": 0, "latency_s": 7.55, "context_chars": 2097, "prompt_chars": 3022, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"National insurance number not found in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "National insurance number not found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "HIST-05", "family": "HIST", "rep": 1, "latency_s": 12.82, "context_chars": 10038, "prompt_chars": 10966, "raw": "```json\n{\"answer\": \"+357 99 100 224\", \"source_id\": \"S-SLF-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Mobile number valid 2019-01-10 through 2024-06-30\"}\n```\n", "obj": {"answer": "+357 99 100 224", "source_id": "S-SLF-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Mobile number valid 2019-01-10 through 2024-06-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HOP-06", "family": "HOP", "rep": 1, "latency_s": 10.57, "context_chars": 4664, "prompt_chars": 5613, "raw": "```json\n{\n  \"answer\": \"Tessera Analytics\",\n  \"source_id\": \"S-DOC-04\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Employed at Tessera Analytics from 2021-12-01 to 2025-02-28.\"\n}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Tessera Analytics from 2021-12-01 to 2025-02-28."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "GAP-03", "family": "GAP", "rep": 1, "latency_s": 6.85, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Trip record exists but contains no cost data\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Trip record exists but contains no cost data"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CONF-01", "family": "CONF", "rep": 1, "latency_s": 20.31, "context_chars": 6744, "prompt_chars": 7664, "raw": "```json\n{\"answer\": \"7 Kalymnou, Larnaca\", \"source_id\": \"S-DOC-03\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Document scan tier 1 wins over self-reported tier 4 per conflict policy.\"}\n```\n", "obj": {"answer": "7 Kalymnou, Larnaca", "source_id": "S-DOC-03", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Document scan tier 1 wins over self-reported tier 4 per conflict policy."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-04", "family": "CURR", "rep": 0, "latency_s": 10.12, "context_chars": 4903, "prompt_chars": 5847, "raw": "```json\n{\"answer\": \"22 euro\", \"source_id\": \"F-084\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Journal subscription active from 2024-03-15, ongoing\"}\n```\n", "obj": {"answer": "22 euro", "source_id": "F-084", "policy_id": null, "unanswerable": false, "stale": false, "note": "Journal subscription active from 2024-03-15, ongoing"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HIST-04", "family": "HIST", "rep": 1, "latency_s": 7.58, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\"answer\": \"Halvard Systems\", \"source_id\": \"S-EML-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed at Halvard Systems from 2018-09-01 to 2021-11-30\"}\n```\n", "obj": {"answer": "Halvard Systems", "source_id": "S-EML-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Halvard Systems from 2018-09-01 to 2021-11-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HIST-06", "family": "HIST", "rep": 0, "latency_s": 8.56, "context_chars": 4903, "prompt_chars": 5822, "raw": "```json\n{\"answer\": \"vegetarian\", \"source_id\": \"F-050\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Diet record F-050 valid 2023-06-02 to 2025-03-29\"}\n```\n", "obj": {"answer": "vegetarian", "source_id": "F-050", "policy_id": null, "unanswerable": false, "stale": false, "note": "Diet record F-050 valid 2023-06-02 to 2025-03-29"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HIST-01", "family": "HIST", "rep": 0, "latency_s": 9.75, "context_chars": 10038, "prompt_chars": 10957, "raw": "```json\n{\n  \"answer\": \"48 Aristotelous, Nicosia\",\n  \"source_id\": \"S-DOC-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Valid from 2022-08-01 to 2025-05-14 per document scan.\"\n}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2022-08-01 to 2025-05-14 per document scan."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HIST-06", "family": "HIST", "rep": 1, "latency_s": 7.95, "context_chars": 2097, "prompt_chars": 3016, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records only contain current diet; no historical data for 2024-01-01\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records only contain current diet; no historical data for 2024-01-01"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "CURR-08", "family": "CURR", "rep": 1, "latency_s": 13.11, "context_chars": 5242, "prompt_chars": 6189, "raw": "```json\n{\n  \"answer\": \"EUR 11\",\n  \"source_id\": \"S-SLF-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Music service costs EUR 11 monthly since 2020-01-01\"\n}\n```\n", "obj": {"answer": "EUR 11", "source_id": "S-SLF-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Music service costs EUR 11 monthly since 2020-01-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CONF-04", "family": "CONF", "rep": 1, "latency_s": 17.32, "context_chars": 6744, "prompt_chars": 7665, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"F-021\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Device sync (tier 2) preferred over self-reported (tier 4) per conflict policy.\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "F-021", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Device sync (tier 2) preferred over self-reported (tier 4) per conflict policy."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CURR-01", "family": "CURR", "rep": 1, "latency_s": 29.86, "context_chars": 6744, "prompt_chars": 7659, "raw": "```json\n{\n  \"answer\": \"Kestrel Data Works\",\n  \"source_id\": \"S-DOC-05\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Document scan from Feb 2025, approximately 15 months old.\"\n}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Document scan from Feb 2025, approximately 15 months old."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "STALE-04", "family": "STALE", "rep": 0, "latency_s": 11.34, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\"answer\": null, \"source_id\": \"F-061\", \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Licence number on record but no expiration date provided\"}\n```\n", "obj": {"answer": null, "source_id": "F-061", "policy_id": null, "unanswerable": true, "stale": false, "note": "Licence number on record but no expiration date provided"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "CURR-04", "family": "CURR", "rep": 0, "latency_s": 10.08, "context_chars": 5242, "prompt_chars": 6186, "raw": "```json\n{\n  \"answer\": \"22\",\n  \"source_id\": \"S-EML-07\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Journal access costs EUR 22 monthly since 2024-03-15.\"\n}\n```\n", "obj": {"answer": "22", "source_id": "S-EML-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Journal access costs EUR 22 monthly since 2024-03-15."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-08", "family": "CURR", "rep": 1, "latency_s": 7.13, "context_chars": 4903, "prompt_chars": 5850, "raw": "```json\n{\"answer\": \"11\", \"source_id\": \"F-085\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Music subscription valid from 2020-01-01 with no end date.\"}\n```\n", "obj": {"answer": "11", "source_id": "F-085", "policy_id": null, "unanswerable": false, "stale": false, "note": "Music subscription valid from 2020-01-01 with no end date."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HIST-03", "family": "HIST", "rep": 0, "latency_s": 7.89, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\n  \"answer\": \"Tessera Analytics\",\n  \"source_id\": \"S-DOC-04\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Employed 2021-12-01 to 2025-02-28\"\n}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed 2021-12-01 to 2025-02-28"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CONF-03", "family": "CONF", "rep": 1, "latency_s": 19.3, "context_chars": 6744, "prompt_chars": 7688, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Records show subscription_cloudline values (9, 12) but don't explicitly state they represent monthly cost in euros.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show subscription_cloudline values (9, 12) but don't explicitly state they represent monthly cost in euros."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "STALE-05", "family": "STALE", "rep": 0, "latency_s": 7.6, "context_chars": 10038, "prompt_chars": 10966, "raw": "```json\n{\n  \"answer\": \"Yes, until 2027-02-20\",\n  \"source_id\": \"F-071\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Aurora 14 laptop warranty remains valid today.\"\n}\n```\n", "obj": {"answer": "Yes, until 2027-02-20", "source_id": "F-071", "policy_id": null, "unanswerable": false, "stale": false, "note": "Aurora 14 laptop warranty remains valid today."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "flat", "item": "CONF-02", "family": "CONF", "rep": 0, "latency_s": 13.51, "context_chars": 2097, "prompt_chars": 3027, "raw": "```json\n{\"answer\": \"4200\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Monthly salary from Kestrel Data Works employment record.\"}\n```\n", "obj": {"answer": "4200", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Monthly salary from Kestrel Data Works employment record."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HOP-03", "family": "HOP", "rep": 0, "latency_s": 8.49, "context_chars": 2097, "prompt_chars": 3044, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Mobile number from records; Vienna trip was 2024-09-17\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Mobile number from records; Vienna trip was 2024-09-17"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "ENT-04", "family": "ENT", "rep": 1, "latency_s": 10.69, "context_chars": 6744, "prompt_chars": 7679, "raw": "```json\n{\n  \"answer\": \"3\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Reiner, Alex Doran, and Alex Kovacs listed in contacts.\"\n}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, and Alex Kovacs listed in contacts."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-03", "family": "CONF", "rep": 1, "latency_s": 31.65, "context_chars": 4664, "prompt_chars": 5608, "raw": "```json\n{\"answer\": \"12\", \"source_id\": \"S-SLF-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Most recent documented cost per note S-SLF-05 from 2022-09-05\"}\n```\n", "obj": {"answer": "12", "source_id": "S-SLF-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent documented cost per note S-SLF-05 from 2022-09-05"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 1, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HOP-03", "family": "HOP", "rep": 1, "latency_s": 51.72, "context_chars": 4903, "prompt_chars": 5850, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"F-021\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Two numbers valid on that date (2024-09-17); selected newer\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "F-021", "policy_id": null, "unanswerable": false, "stale": false, "note": "Two numbers valid on that date (2024-09-17); selected newer"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-04", "family": "CURR", "rep": 1, "latency_s": 7.03, "context_chars": 10038, "prompt_chars": 10982, "raw": "```json\n{\"answer\": \"22\", \"source_id\": \"S-EML-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Journal subscription active since 2024-03-15, valid to open.\"}\n```\n", "obj": {"answer": "22", "source_id": "S-EML-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Journal subscription active since 2024-03-15, valid to open."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CONF-05", "family": "CONF", "rep": 0, "latency_s": 15.01, "context_chars": 5242, "prompt_chars": 6152, "raw": "```json\n{\"answer\": \"0+\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Laboratory report from 2024-11-09 confirms 0+ blood type\"}\n```\n", "obj": {"answer": "0+", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Laboratory report from 2024-11-09 confirms 0+ blood type"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "ENT-01", "family": "ENT", "rep": 0, "latency_s": 7.63, "context_chars": 2097, "prompt_chars": 3041, "raw": "```json\n{\"answer\": \"+357 99 018 442\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Kovacs, landlord of 7 Kalymnou in Larnaca.\"}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs, landlord of 7 Kalymnou in Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-02", "family": "CURR", "rep": 1, "latency_s": 8.52, "context_chars": 4664, "prompt_chars": 5576, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": \"S-SLF-08\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Pescatarian since 2025-03-30\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": "S-SLF-08", "policy_id": null, "unanswerable": false, "stale": false, "note": "Pescatarian since 2025-03-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-02", "family": "HIST", "rep": 0, "latency_s": 9.03, "context_chars": 6744, "prompt_chars": 7663, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No home address records available before 2025.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No home address records available before 2025."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CURR-02", "family": "CURR", "rep": 0, "latency_s": 14.58, "context_chars": 6744, "prompt_chars": 7656, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": \"F-051\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Self-reported diet preference captured 2025-03-30.\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": "F-051", "policy_id": null, "unanswerable": false, "stale": false, "note": "Self-reported diet preference captured 2025-03-30."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HOP-01", "family": "HOP", "rep": 0, "latency_s": 9.14, "context_chars": 2097, "prompt_chars": 3055, "raw": "```json\n{\"answer\": \"Nadja Lind\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Most recent trip to Nicosia (sister's city) on 2026-01-22.\"}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent trip to Nicosia (sister's city) on 2026-01-22."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "STALE-02", "family": "STALE", "rep": 0, "latency_s": 20.89, "context_chars": 4903, "prompt_chars": 5826, "raw": "```json\n{\n  \"answer\": \"No, it expired on 2026-08-31.\",\n  \"source_id\": \"F-062\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Health insurance policy HI-88231 validity ended 2026-08-31.\"\n}\n```\n", "obj": {"answer": "No, it expired on 2026-08-31.", "source_id": "F-062", "policy_id": null, "unanswerable": false, "stale": true, "note": "Health insurance policy HI-88231 validity ended 2026-08-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "flat", "item": "GAP-05", "family": "GAP", "rep": 1, "latency_s": 8.07, "context_chars": 2097, "prompt_chars": 3036, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Employer name is Kestrel Data Works, but office address not recorded\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Employer name is Kestrel Data Works, but office address not recorded"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "GAP-05", "family": "GAP", "rep": 0, "latency_s": 9.14, "context_chars": 4903, "prompt_chars": 5842, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"No employer office street address found in records.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No employer office street address found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "ENT-02", "family": "ENT", "rep": 0, "latency_s": 13.53, "context_chars": 4664, "prompt_chars": 5595, "raw": "```json\n{\"answer\": \"Alex Doran, in Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran listed as GP in Larnaca\"}\n```\n", "obj": {"answer": "Alex Doran, in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran listed as GP in Larnaca"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HOP-04", "family": "HOP", "rep": 0, "latency_s": 47.58, "context_chars": 4664, "prompt_chars": 5643, "raw": "```json\n{\n  \"answer\": \"Ilona Vantaa and Nadja Lind\",\n  \"source_id\": \"S-DOC-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Both live in Nicosia; user lived there 2022-2025.\"\n}\n```\n", "obj": {"answer": "Ilona Vantaa and Nadja Lind", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Both live in Nicosia; user lived there 2022-2025."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "GAP-01", "family": "GAP", "rep": 1, "latency_s": 8.12, "context_chars": 4903, "prompt_chars": 5828, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"National insurance number not recorded.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "National insurance number not recorded."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HOP-06", "family": "HOP", "rep": 0, "latency_s": 10.59, "context_chars": 4664, "prompt_chars": 5613, "raw": "```json\n{\"answer\": \"Tessera Analytics\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed at Tessera Analytics from 2021-12-01 to 2025-02-28\"}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Tessera Analytics from 2021-12-01 to 2025-02-28"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HOP-05", "family": "HOP", "rep": 1, "latency_s": 61.29, "context_chars": 5242, "prompt_chars": 6198, "raw": "```json\n{\n  \"answer\": \"99\",\n  \"source_id\": \"S-SLF-02, S-SLF-05, S-EML-06, S-EML-05, S-EML-07\",\n  \"policy_id\": \"S-SLF-05 over S-EML-04\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Music 11 + Cloudline 12 + Gym 35 + Trainline 19 + Journal 22\"\n}\n```\n", "obj": {"answer": "99", "source_id": "S-SLF-02, S-SLF-05, S-EML-06, S-EML-05, S-EML-07", "policy_id": "S-SLF-05 over S-EML-04", "unanswerable": false, "stale": false, "note": "Music 11 + Cloudline 12 + Gym 35 + Trainline 19 + Journal 22"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "ENT-04", "family": "ENT", "rep": 0, "latency_s": 9.87, "context_chars": 4903, "prompt_chars": 5838, "raw": "```json\n{\n  \"answer\": \"3\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Reiner, Alex Doran, and Alex Kovacs in contacts\"\n}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, and Alex Kovacs in contacts"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CONF-03", "family": "CONF", "rep": 1, "latency_s": 14.46, "context_chars": 10038, "prompt_chars": 10982, "raw": "```json\n{\"answer\": \"9\", \"source_id\": \"S-EML-04\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Conflicting values resolved by lower source tier (email vs self-reported).\"}\n```\n", "obj": {"answer": "9", "source_id": "S-EML-04", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Conflicting values resolved by lower source tier (email vs self-reported)."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "STALE-01", "family": "STALE", "rep": 1, "latency_s": 11.33, "context_chars": 4664, "prompt_chars": 5641, "raw": "```json\n{\"answer\": \"Passport P4417832 expired 2026-04-11. Cannot use it.\", \"source_id\": \"S-DOC-06\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Passport expired 148 days ago.\"}\n```\n", "obj": {"answer": "Passport P4417832 expired 2026-04-11. Cannot use it.", "source_id": "S-DOC-06", "policy_id": null, "unanswerable": false, "stale": false, "note": "Passport expired 148 days ago."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "ENT-02", "family": "ENT", "rep": 0, "latency_s": 8.33, "context_chars": 5242, "prompt_chars": 6173, "raw": "```json\n{\"answer\": \"Alex Doran, in Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran is your GP in Larnaca.\"}\n```\n", "obj": {"answer": "Alex Doran, in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran is your GP in Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HOP-05", "family": "HOP", "rep": 0, "latency_s": 9.36, "context_chars": 2097, "prompt_chars": 3053, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records show subscription IDs but not costs or active dates.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show subscription IDs but not costs or active dates."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "CONF-06", "family": "CONF", "rep": 0, "latency_s": 8.14, "context_chars": 2097, "prompt_chars": 3063, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Only one salary record found in provided records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Only one salary record found in provided records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HIST-03", "family": "HIST", "rep": 1, "latency_s": 7.81, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\n  \"answer\": \"Tessera Analytics\",\n  \"source_id\": \"S-DOC-04\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Employed at Tessera Analytics from 2021-12-01 to 2025-02-28.\"\n}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Tessera Analytics from 2021-12-01 to 2025-02-28."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HIST-06", "family": "HIST", "rep": 0, "latency_s": 10.24, "context_chars": 2097, "prompt_chars": 3016, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records show current diet only; no historical data for 2024-01-01.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show current diet only; no historical data for 2024-01-01."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "CURR-02", "family": "CURR", "rep": 0, "latency_s": 9.31, "context_chars": 2097, "prompt_chars": 3009, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Diet recorded in personal facts.\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Diet recorded in personal facts."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HOP-04", "family": "HOP", "rep": 1, "latency_s": 36.99, "context_chars": 4903, "prompt_chars": 5882, "raw": "```json\n{\n  \"answer\": \"Ilona Vantaa and Nadja Lind\",\n  \"source_id\": \"F-002\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Previous address in Nicosia before current address.\"\n}\n```\n", "obj": {"answer": "Ilona Vantaa and Nadja Lind", "source_id": "F-002", "policy_id": null, "unanswerable": false, "stale": false, "note": "Previous address in Nicosia before current address."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "GAP-04", "family": "GAP", "rep": 0, "latency_s": 9.8, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain only Ilona Vantaa's phone number, not email address.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain only Ilona Vantaa's phone number, not email address."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HOP-03", "family": "HOP", "rep": 0, "latency_s": 9.15, "context_chars": 5242, "prompt_chars": 6189, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": \"S-DEV-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Vienna trip 2024-09-17; new number active from 2024-07-01\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Vienna trip 2024-09-17; new number active from 2024-07-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "ENT-01", "family": "ENT", "rep": 0, "latency_s": 10.31, "context_chars": 5242, "prompt_chars": 6186, "raw": "```json\n{\"answer\": \"+357 99 018 442\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Kovacs is your landlord at 7 Kalymnou, Larnaca.\"}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs is your landlord at 7 Kalymnou, Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "STALE-01", "family": "STALE", "rep": 0, "latency_s": 22.36, "context_chars": 5242, "prompt_chars": 6219, "raw": "```json\n{\n  \"answer\": \"Passport P4417832 expired 2026-04-11; cannot be used\",\n  \"source_id\": \"S-DOC-06\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Passport expired before today. Renewal required for travel.\"\n}\n```\n", "obj": {"answer": "Passport P4417832 expired 2026-04-11; cannot be used", "source_id": "S-DOC-06", "policy_id": null, "unanswerable": false, "stale": false, "note": "Passport expired before today. Renewal required for travel."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "flat", "item": "CONF-04", "family": "CONF", "rep": 0, "latency_s": 6.79, "context_chars": 2097, "prompt_chars": 3018, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current mobile phone number from records\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Current mobile phone number from records"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "GAP-01", "family": "GAP", "rep": 1, "latency_s": 7.77, "context_chars": 5242, "prompt_chars": 6167, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"National insurance number not found in records.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "National insurance number not found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "CONF-03", "family": "CONF", "rep": 0, "latency_s": 8.38, "context_chars": 2097, "prompt_chars": 3041, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Record shows subscription_cloudline value of 12 but cost per month not specified.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Record shows subscription_cloudline value of 12 but cost per month not specified."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "CURR-04", "family": "CURR", "rep": 0, "latency_s": 14.6, "context_chars": 2097, "prompt_chars": 3041, "raw": "```json\n{\"answer\": \"22 EUR\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Journal subscription costs 22 EUR per month.\"}\n```\n", "obj": {"answer": "22 EUR", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Journal subscription costs 22 EUR per month."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "ENT-04", "family": "ENT", "rep": 1, "latency_s": 7.64, "context_chars": 2097, "prompt_chars": 3032, "raw": "```json\n{\"answer\": \"3\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, Alex Doran, and Alex Kovacs are listed in contacts.\"}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, and Alex Kovacs are listed in contacts."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-05", "family": "CURR", "rep": 1, "latency_s": 8.45, "context_chars": 4664, "prompt_chars": 5574, "raw": "```json\n{\"answer\": \"Aurora 14\", \"source_id\": \"S-EML-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Purchased 2024-02-20, warranty valid until 2027-02-20\"}\n```\n", "obj": {"answer": "Aurora 14", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Purchased 2024-02-20, warranty valid until 2027-02-20"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "GAP-04", "family": "GAP", "rep": 0, "latency_s": 6.97, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister's email address not in records\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister's email address not in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "GAP-04", "family": "GAP", "rep": 0, "latency_s": 7.77, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister identified but email address not in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister identified but email address not in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "STALE-01", "family": "STALE", "rep": 0, "latency_s": 7.6, "context_chars": 4664, "prompt_chars": 5641, "raw": "```json\n{\n  \"answer\": \"Passport P4417832 expired 2026-04-11; cannot be used today.\",\n  \"source_id\": \"S-DOC-06\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Passport expired on 2026-04-11, before today 2026-09-06.\"\n}\n```\n", "obj": {"answer": "Passport P4417832 expired 2026-04-11; cannot be used today.", "source_id": "S-DOC-06", "policy_id": null, "unanswerable": false, "stale": true, "note": "Passport expired on 2026-04-11, before today 2026-09-06."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes", "item": "HIST-08", "family": "HIST", "rep": 1, "latency_s": 10.07, "context_chars": 4664, "prompt_chars": 5617, "raw": "```json\n{\"answer\": \"EUR 19\", \"source_id\": \"S-EML-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Trainline pass cost EUR 19/month from 2023-01-01 to 2024-12-31.\"}\n```\n", "obj": {"answer": "EUR 19", "source_id": "S-EML-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Trainline pass cost EUR 19/month from 2023-01-01 to 2024-12-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "ENT-01", "family": "ENT", "rep": 1, "latency_s": 10.27, "context_chars": 6744, "prompt_chars": 7688, "raw": "```json\n{\n  \"answer\": \"+357 99 018 442\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Kovacs is your landlord at 7 Kalymnou, Larnaca.\"\n}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs is your landlord at 7 Kalymnou, Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "ENT-05", "family": "ENT", "rep": 1, "latency_s": 19.05, "context_chars": 5242, "prompt_chars": 6189, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Colleague at Kestrel Data Works, current employer since 2025-03-01.\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Colleague at Kestrel Data Works, current employer since 2025-03-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-06", "family": "CURR", "rep": 1, "latency_s": 8.34, "context_chars": 10038, "prompt_chars": 10947, "raw": "```json\n{\n  \"answer\": \"CY17 0020 0128 0000 0012 0052 7600\",\n  \"source_id\": \"S-DOC-04\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Bank IBAN from document scan, valid from 2021-12-01 to present\"\n}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Bank IBAN from document scan, valid from 2021-12-01 to present"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-08", "family": "CURR", "rep": 0, "latency_s": 10.26, "context_chars": 10038, "prompt_chars": 10985, "raw": "```json\n{\"answer\": \"11\", \"source_id\": \"F-085\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Music subscription costs 11 EUR/month, valid since 2020-01-01\"}\n```\n", "obj": {"answer": "11", "source_id": "F-085", "policy_id": null, "unanswerable": false, "stale": false, "note": "Music subscription costs 11 EUR/month, valid since 2020-01-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HOP-02", "family": "HOP", "rep": 0, "latency_s": 11.97, "context_chars": 5242, "prompt_chars": 6193, "raw": "```json\n{\"answer\": \"Limassol\", \"source_id\": \"S-DOC-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Tessera Analytics started 2021-12-01; Limassol lease 2019-03-01 to 2022-07-31.\"}\n```\n", "obj": {"answer": "Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Tessera Analytics started 2021-12-01; Limassol lease 2019-03-01 to 2022-07-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-08", "family": "HIST", "rep": 0, "latency_s": 8.37, "context_chars": 6744, "prompt_chars": 7697, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No Trainline pass subscription recorded in available facts.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No Trainline pass subscription recorded in available facts."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "HIST-02", "family": "HIST", "rep": 0, "latency_s": 9.26, "context_chars": 10038, "prompt_chars": 10957, "raw": "```json\n{\"answer\": \"12 Harbour Lane, Limassol\", \"source_id\": \"F-001\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid address period: 2019-03-01 to 2022-07-31\"}\n```\n", "obj": {"answer": "12 Harbour Lane, Limassol", "source_id": "F-001", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid address period: 2019-03-01 to 2022-07-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HIST-05", "family": "HIST", "rep": 0, "latency_s": 10.21, "context_chars": 4903, "prompt_chars": 5831, "raw": "```json\n{\"answer\": \"+357 99 100 224\", \"source_id\": \"F-020\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2019-01-10 to 2024-06-30; active on 2023-03-01\"}\n```\n", "obj": {"answer": "+357 99 100 224", "source_id": "F-020", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2019-01-10 to 2024-06-30; active on 2023-03-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HOP-05", "family": "HOP", "rep": 1, "latency_s": 9.06, "context_chars": 2097, "prompt_chars": 3053, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records show current subscriptions but lack historical data for 2024-06-01\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show current subscriptions but lack historical data for 2024-06-01"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "CURR-08", "family": "CURR", "rep": 0, "latency_s": 8.02, "context_chars": 4903, "prompt_chars": 5850, "raw": "```json\n{\"answer\": \"11\", \"source_id\": \"F-085\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Music subscription costs €11/month, active since 2020-01-01\"}\n```\n", "obj": {"answer": "11", "source_id": "F-085", "policy_id": null, "unanswerable": false, "stale": false, "note": "Music subscription costs €11/month, active since 2020-01-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "STALE-01", "family": "STALE", "rep": 0, "latency_s": 6.78, "context_chars": 2097, "prompt_chars": 3074, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Passport number not found in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Passport number not found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CURR-03", "family": "CURR", "rep": 1, "latency_s": 9.21, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Document scan from 2017-02-03\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Document scan from 2017-02-03"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "ENT-04", "family": "ENT", "rep": 0, "latency_s": 8.19, "context_chars": 10038, "prompt_chars": 10973, "raw": "```json\n{\"answer\": \"3\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, Alex Doran, and Alex Kovacs in contacts\"}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, and Alex Kovacs in contacts"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CONF-01", "family": "CONF", "rep": 1, "latency_s": 8.32, "context_chars": 2097, "prompt_chars": 3017, "raw": "```json\n{\"answer\": \"48 Aristotelous, Nicosia\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current home address from personal records\"}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Current home address from personal records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-05", "family": "CONF", "rep": 1, "latency_s": 15.82, "context_chars": 4664, "prompt_chars": 5574, "raw": "```json\n{\"answer\": \"0+\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Laboratory report from 2024-11-09 is more authoritative than earlier personal note\"}\n```\n", "obj": {"answer": "0+", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Laboratory report from 2024-11-09 is more authoritative than earlier personal note"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HIST-02", "family": "HIST", "rep": 0, "latency_s": 8.69, "context_chars": 5242, "prompt_chars": 6161, "raw": "```json\n{\"answer\": \"12 Harbour Lane, Limassol\", \"source_id\": \"S-DOC-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Lived there from 2019-03-01 to 2022-07-31\"}\n```\n", "obj": {"answer": "12 Harbour Lane, Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Lived there from 2019-03-01 to 2022-07-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HOP-01", "family": "HOP", "rep": 0, "latency_s": 8.57, "context_chars": 4903, "prompt_chars": 5861, "raw": "```json\n{\n  \"answer\": \"Nadja Lind\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Sister Ilona lives in Nicosia. Most recent Nicosia trip was 2026-01-22.\"\n}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Sister Ilona lives in Nicosia. Most recent Nicosia trip was 2026-01-22."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CONF-01", "family": "CONF", "rep": 1, "latency_s": 30.13, "context_chars": 4903, "prompt_chars": 5823, "raw": "```json\n{\"answer\": \"7 Kalymnou, Larnaca\", \"source_id\": \"F-003\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Two valid addresses from 2025-05-15; F-003 selected as primary.\"}\n```\n", "obj": {"answer": "7 Kalymnou, Larnaca", "source_id": "F-003", "policy_id": null, "unanswerable": false, "stale": false, "note": "Two valid addresses from 2025-05-15; F-003 selected as primary."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-08", "family": "CURR", "rep": 1, "latency_s": 14.15, "context_chars": 4664, "prompt_chars": 5611, "raw": "```json\n{\"answer\": \"11\", \"source_id\": \"S-SLF-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Music service costs EUR 11 per month since 2020-01-01\"}\n```\n", "obj": {"answer": "11", "source_id": "S-SLF-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Music service costs EUR 11 per month since 2020-01-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "GAP-01", "family": "GAP", "rep": 0, "latency_s": 8.05, "context_chars": 10038, "prompt_chars": 10963, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"National insurance number not found in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "National insurance number not found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "CURR-02", "family": "CURR", "rep": 0, "latency_s": 7.06, "context_chars": 4664, "prompt_chars": 5576, "raw": "```json\n{\n  \"answer\": \"pescatarian\",\n  \"source_id\": \"S-SLF-08\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Pescatarian since 2025-03-30\"\n}\n```\n", "obj": {"answer": "pescatarian", "source_id": "S-SLF-08", "policy_id": null, "unanswerable": false, "stale": false, "note": "Pescatarian since 2025-03-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-01", "family": "CURR", "rep": 1, "latency_s": 8.73, "context_chars": 2097, "prompt_chars": 3012, "raw": "```json\n{\"answer\": \"Kestrel Data Works\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Listed in facts section as current employer\"}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Listed in facts section as current employer"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "ENT-03", "family": "ENT", "rep": 0, "latency_s": 9.1, "context_chars": 4903, "prompt_chars": 5855, "raw": "```json\n{\n  \"answer\": \"+357 96 220 145\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Reiner accompanied Mira to Berlin on 2026-03-14.\"\n}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner accompanied Mira to Berlin on 2026-03-14."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "STALE-05", "family": "STALE", "rep": 1, "latency_s": 10.43, "context_chars": 5242, "prompt_chars": 6170, "raw": "```json\n{\n  \"answer\": \"Yes, your Aurora 14 laptop warranty extends to 2027-02-20.\",\n  \"source_id\": \"S-EML-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Warranty period 2024-02-20 to 2027-02-20 remains valid.\"\n}\n```\n", "obj": {"answer": "Yes, your Aurora 14 laptop warranty extends to 2027-02-20.", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Warranty period 2024-02-20 to 2027-02-20 remains valid."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dimension", "item": "HOP-02", "family": "HOP", "rep": 0, "latency_s": 10.13, "context_chars": 10038, "prompt_chars": 10989, "raw": "```json\n{\"answer\": \"Limassol\", \"source_id\": \"F-001\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Tessera Analytics employment started 2021-12-01; home address 12 Harbour Lane, Limassol valid through 2022-07-31.\"}\n```\n", "obj": {"answer": "Limassol", "source_id": "F-001", "policy_id": null, "unanswerable": false, "stale": false, "note": "Tessera Analytics employment started 2021-12-01; home address 12 Harbour Lane, Limassol valid through 2022-07-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HOP-03", "family": "HOP", "rep": 1, "latency_s": 14.34, "context_chars": 2097, "prompt_chars": 3044, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"From records; Vienna trip confirmed 2024-09-17\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "From records; Vienna trip confirmed 2024-09-17"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "GAP-04", "family": "GAP", "rep": 0, "latency_s": 7.81, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister Ilona Vantaa identified, but email address not in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister Ilona Vantaa identified, but email address not in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "HIST-07", "family": "HIST", "rep": 1, "latency_s": 7.35, "context_chars": 10038, "prompt_chars": 10993, "raw": "```json\n{\"answer\": \"35\", \"source_id\": \"S-EML-06\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Gym subscription €35/month valid 2022-09-01 to 2025-05-31\"}\n```\n", "obj": {"answer": "35", "source_id": "S-EML-06", "policy_id": null, "unanswerable": false, "stale": false, "note": "Gym subscription €35/month valid 2022-09-01 to 2025-05-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CURR-06", "family": "CURR", "rep": 1, "latency_s": 13.71, "context_chars": 6744, "prompt_chars": 7653, "raw": "```json\n{\"answer\": \"CY17 0020 0128 0000 0012 0052 7600\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Document scan from 2021-11-18\"}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Document scan from 2021-11-18"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-01", "family": "CURR", "rep": 0, "latency_s": 10.09, "context_chars": 2097, "prompt_chars": 3012, "raw": "```json\n{\"answer\": \"Kestrel Data Works\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current employer from personal records\"}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Current employer from personal records"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CONF-04", "family": "CONF", "rep": 0, "latency_s": 24.37, "context_chars": 4903, "prompt_chars": 5824, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"F-021\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Two numbers currently valid since 2024-07-01; also +357 99 100 224\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "F-021", "policy_id": null, "unanswerable": false, "stale": false, "note": "Two numbers currently valid since 2024-07-01; also +357 99 100 224"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "STALE-02", "family": "STALE", "rep": 1, "latency_s": 37.56, "context_chars": 6744, "prompt_chars": 7667, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": \"S-DOC-08\",\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": true,\n  \"note\": \"Health insurance review was due 2026-08-31, now overdue; current status unknown.\"\n}\n```\n", "obj": {"answer": null, "source_id": "S-DOC-08", "policy_id": null, "unanswerable": true, "stale": true, "note": "Health insurance review was due 2026-08-31, now overdue; current status unknown."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "STALE-03", "family": "STALE", "rep": 1, "latency_s": 14.14, "context_chars": 4903, "prompt_chars": 5832, "raw": "```json\n{\n  \"answer\": \"No, the warranty expired on 2026-07-01.\",\n  \"source_id\": \"F-073\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Corvus 8 phone warranty period ended over 2 months ago.\"\n}\n```\n", "obj": {"answer": "No, the warranty expired on 2026-07-01.", "source_id": "F-073", "policy_id": null, "unanswerable": false, "stale": false, "note": "Corvus 8 phone warranty period ended over 2 months ago."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "ENT-02", "family": "ENT", "rep": 1, "latency_s": 10.65, "context_chars": 6744, "prompt_chars": 7675, "raw": "```json\n{\"answer\": \"Alex Doran, in Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran listed as general practitioner in people records.\"}\n```\n", "obj": {"answer": "Alex Doran, in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran listed as general practitioner in people records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HIST-07", "family": "HIST", "rep": 0, "latency_s": 9.63, "context_chars": 10038, "prompt_chars": 10993, "raw": "```json\n{\"answer\": \"35\", \"source_id\": \"S-EML-06\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Gym subscription 35 EUR/month valid 2022-09-01 to 2025-05-31\"}\n```\n", "obj": {"answer": "35", "source_id": "S-EML-06", "policy_id": null, "unanswerable": false, "stale": false, "note": "Gym subscription 35 EUR/month valid 2022-09-01 to 2025-05-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-05", "family": "CONF", "rep": 0, "latency_s": 17.53, "context_chars": 4664, "prompt_chars": 5574, "raw": "```json\n{\"answer\": \"0+\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Laboratory report (2024-11-09) conflicts with earlier personal note; laboratory report used.\"}\n```\n", "obj": {"answer": "0+", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Laboratory report (2024-11-09) conflicts with earlier personal note; laboratory report used."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "GAP-05", "family": "GAP", "rep": 0, "latency_s": 10.84, "context_chars": 5242, "prompt_chars": 6181, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records identify employer but contain no office street address.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records identify employer but contain no office street address."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "CONF-05", "family": "CONF", "rep": 1, "latency_s": 19.22, "context_chars": 5242, "prompt_chars": 6152, "raw": "```json\n{\"answer\": \"0+\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Laboratory report 2024-11-09 supersedes earlier personal note from 2023.\"}\n```\n", "obj": {"answer": "0+", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Laboratory report 2024-11-09 supersedes earlier personal note from 2023."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HOP-04", "family": "HOP", "rep": 1, "latency_s": 12.6, "context_chars": 4664, "prompt_chars": 5643, "raw": "```json\n{\n  \"answer\": \"Ilona Vantaa, Nadja Lind\",\n  \"source_id\": \"S-DOC-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Previous address 48 Aristotelous, Nicosia (until 2025-05-14); two contacts reside there.\"\n}\n```\n", "obj": {"answer": "Ilona Vantaa, Nadja Lind", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Previous address 48 Aristotelous, Nicosia (until 2025-05-14); two contacts reside there."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "GAP-05", "family": "GAP", "rep": 1, "latency_s": 8.79, "context_chars": 4903, "prompt_chars": 5842, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain employer names only, not office addresses.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain employer names only, not office addresses."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HOP-02", "family": "HOP", "rep": 1, "latency_s": 10.2, "context_chars": 6744, "prompt_chars": 7695, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Records contain no employment start date for Tessera Analytics or historical address data.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain no employment start date for Tessera Analytics or historical address data."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HIST-06", "family": "HIST", "rep": 0, "latency_s": 8.38, "context_chars": 4664, "prompt_chars": 5583, "raw": "```json\n{\n  \"answer\": \"vegetarian\",\n  \"source_id\": \"S-SLF-03\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vegetarian period: 2023-06-02 to 2025-03-29, includes 2024-01-01.\"\n}\n```\n", "obj": {"answer": "vegetarian", "source_id": "S-SLF-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Vegetarian period: 2023-06-02 to 2025-03-29, includes 2024-01-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HOP-01", "family": "HOP", "rep": 1, "latency_s": 11.88, "context_chars": 10038, "prompt_chars": 10996, "raw": "```json\n{\n  \"answer\": \"Nadja Lind\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Most recent Nicosia trip (2026-01-22) with Nadja Lind. Sister Ilona Vantaa lives in Nicosia.\"\n}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent Nicosia trip (2026-01-22) with Nadja Lind. Sister Ilona Vantaa lives in Nicosia."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "ENT-03", "family": "ENT", "rep": 1, "latency_s": 11.1, "context_chars": 5242, "prompt_chars": 6194, "raw": "```json\n{\"answer\": \"+357 96 220 145\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, travelled to Berlin 2026-03-14\"}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, travelled to Berlin 2026-03-14"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "GAP-02", "family": "GAP", "rep": 0, "latency_s": 7.68, "context_chars": 4903, "prompt_chars": 5816, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"No home address records cover 2016; earliest record starts 2019.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No home address records cover 2016; earliest record starts 2019."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "ENT-01", "family": "ENT", "rep": 0, "latency_s": 9.03, "context_chars": 10038, "prompt_chars": 10982, "raw": "```json\n{\"answer\": \"+357 99 018 442\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Kovacs is your landlord at 7 Kalymnou, Larnaca.\"}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs is your landlord at 7 Kalymnou, Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "STALE-01", "family": "STALE", "rep": 1, "latency_s": 7.57, "context_chars": 2097, "prompt_chars": 3074, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Passport number not found in records for Mira Vantaa.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Passport number not found in records for Mira Vantaa."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HIST-07", "family": "HIST", "rep": 0, "latency_s": 8.24, "context_chars": 4903, "prompt_chars": 5858, "raw": "```json\n{\"answer\": \"35\", \"source_id\": \"F-083\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Gym subscription active on 2024-06-01\"}\n```\n", "obj": {"answer": "35", "source_id": "F-083", "policy_id": null, "unanswerable": false, "stale": false, "note": "Gym subscription active on 2024-06-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HOP-02", "family": "HOP", "rep": 1, "latency_s": 11.67, "context_chars": 5242, "prompt_chars": 6193, "raw": "```json\n{\n  \"answer\": \"Limassol\",\n  \"source_id\": \"S-DOC-01\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Started Tessera Analytics 2021-12-01; lived at 12 Harbour Lane, Limassol until 2022-07-31.\"\n}\n```\n", "obj": {"answer": "Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Started Tessera Analytics 2021-12-01; lived at 12 Harbour Lane, Limassol until 2022-07-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CONF-03", "family": "CONF", "rep": 1, "latency_s": 12.14, "context_chars": 2097, "prompt_chars": 3041, "raw": "```json\n{\n  \"answer\": \"12\",\n  \"source_id\": \"subscription_cloudline\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Cloudline subscription value listed as 12; currency and monthly period inferred from context.\"\n}\n```\n", "obj": {"answer": "12", "source_id": "subscription_cloudline", "policy_id": null, "unanswerable": false, "stale": false, "note": "Cloudline subscription value listed as 12; currency and monthly period inferred from context."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HIST-01", "family": "HIST", "rep": 1, "latency_s": 9.45, "context_chars": 4903, "prompt_chars": 5822, "raw": "```json\n{\n  \"answer\": \"48 Aristotelous, Nicosia\",\n  \"source_id\": \"F-002\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Valid address from 2022-08-01 to 2025-05-14\"\n}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": "F-002", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid address from 2022-08-01 to 2025-05-14"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CONF-04", "family": "CONF", "rep": 1, "latency_s": 7.62, "context_chars": 2097, "prompt_chars": 3018, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Mobile phone attribute from your personal records.\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Mobile phone attribute from your personal records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CONF-02", "family": "CONF", "rep": 0, "latency_s": 55.76, "context_chars": 5242, "prompt_chars": 6172, "raw": "```json\n{\"answer\": \"4410\", \"source_id\": \"S-DOC-09\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Payslip shows EUR 4410; own note contradicts with EUR 4200\"}\n```\n", "obj": {"answer": "4410", "source_id": "S-DOC-09", "policy_id": null, "unanswerable": false, "stale": false, "note": "Payslip shows EUR 4410; own note contradicts with EUR 4200"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-03", "family": "HIST", "rep": 0, "latency_s": 17.9, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Only employer fact available is from 2025 document; no explicit 2022-05-01 data.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Only employer fact available is from 2025 document; no explicit 2022-05-01 data."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "HOP-03", "family": "HOP", "rep": 1, "latency_s": 11.88, "context_chars": 10038, "prompt_chars": 10985, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"F-021\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vienna trip 2024-09-17. Conflict resolved: device_sync (tier 2) prevails over self_reported (tier 4).\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "F-021", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Vienna trip 2024-09-17. Conflict resolved: device_sync (tier 2) prevails over self_reported (tier 4)."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-07", "family": "CURR", "rep": 1, "latency_s": 7.39, "context_chars": 4903, "prompt_chars": 5819, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": \"F-064\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Drug allergy record currently valid\"}\n```\n", "obj": {"answer": "penicillin", "source_id": "F-064", "policy_id": null, "unanswerable": false, "stale": false, "note": "Drug allergy record currently valid"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "ENT-02", "family": "ENT", "rep": 1, "latency_s": 10.25, "context_chars": 10038, "prompt_chars": 10969, "raw": "```json\n{\"answer\": \"Alex Doran, in Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran is listed as general practitioner in Larnaca.\"}\n```\n", "obj": {"answer": "Alex Doran, in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran is listed as general practitioner in Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HOP-04", "family": "HOP", "rep": 0, "latency_s": 24.27, "context_chars": 4903, "prompt_chars": 5882, "raw": "```json\n{\n  \"answer\": \"Ilona Vantaa, Nadja Lind\",\n  \"source_id\": \"F-002\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Previous address was 48 Aristotelous, Nicosia; two contacts found there.\"\n}\n```\n", "obj": {"answer": "Ilona Vantaa, Nadja Lind", "source_id": "F-002", "policy_id": null, "unanswerable": false, "stale": false, "note": "Previous address was 48 Aristotelous, Nicosia; two contacts found there."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CONF-05", "family": "CONF", "rep": 0, "latency_s": 7.5, "context_chars": 2097, "prompt_chars": 3007, "raw": "```json\n{\"answer\": \"0+\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Blood type from personal health records.\"}\n```\n", "obj": {"answer": "0+", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Blood type from personal health records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "STALE-03", "family": "STALE", "rep": 0, "latency_s": 9.15, "context_chars": 6744, "prompt_chars": 7673, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Handset (Corvus 8) identified but warranty data not recorded.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Handset (Corvus 8) identified but warranty data not recorded."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "CURR-02", "family": "CURR", "rep": 1, "latency_s": 9.47, "context_chars": 2097, "prompt_chars": 3009, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Direct record from personal facts\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Direct record from personal facts"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "STALE-03", "family": "STALE", "rep": 0, "latency_s": 10.9, "context_chars": 4903, "prompt_chars": 5832, "raw": "```json\n{\n  \"answer\": \"No, warranty expired on 2026-07-01\",\n  \"source_id\": \"F-073\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Phone warranty expired 67 days ago\"\n}\n```\n", "obj": {"answer": "No, warranty expired on 2026-07-01", "source_id": "F-073", "policy_id": null, "unanswerable": false, "stale": true, "note": "Phone warranty expired 67 days ago"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes", "item": "GAP-05", "family": "GAP", "rep": 0, "latency_s": 8.54, "context_chars": 4664, "prompt_chars": 5603, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No employer office address found in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No employer office address found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "CONF-06", "family": "CONF", "rep": 1, "latency_s": 14.73, "context_chars": 4903, "prompt_chars": 5869, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Records F-031 (€4410) and F-032 (€4200) both claim validity from 2026-01-01 onward. No conflict resolution policy provided.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records F-031 (€4410) and F-032 (€4200) both claim validity from 2026-01-01 onward. No conflict resolution policy provided."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HIST-06", "family": "HIST", "rep": 0, "latency_s": 9.33, "context_chars": 6744, "prompt_chars": 7663, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Only current diet (pescatarian as of 2025-03-30) is recorded; past diet unknown.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Only current diet (pescatarian as of 2025-03-30) is recorded; past diet unknown."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "HIST-06", "family": "HIST", "rep": 1, "latency_s": 8.01, "context_chars": 10038, "prompt_chars": 10957, "raw": "```json\n{\n  \"answer\": \"vegetarian\",\n  \"source_id\": \"F-050\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Diet was vegetarian from 2023-06-02 through 2025-03-29.\"\n}\n```\n", "obj": {"answer": "vegetarian", "source_id": "F-050", "policy_id": null, "unanswerable": false, "stale": false, "note": "Diet was vegetarian from 2023-06-02 through 2025-03-29."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HIST-03", "family": "HIST", "rep": 0, "latency_s": 10.4, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\"answer\": \"Tessera Analytics\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed at Tessera Analytics from 2021-12-01 to 2025-02-28.\"}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Tessera Analytics from 2021-12-01 to 2025-02-28."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-05", "family": "CURR", "rep": 1, "latency_s": 7.46, "context_chars": 4903, "prompt_chars": 5813, "raw": "```json\n{\"answer\": \"Aurora 14\", \"source_id\": \"F-070\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Aurora 14 laptop purchased 2024-02-20, warranty until 2027-02-20\"}\n```\n", "obj": {"answer": "Aurora 14", "source_id": "F-070", "policy_id": null, "unanswerable": false, "stale": false, "note": "Aurora 14 laptop purchased 2024-02-20, warranty until 2027-02-20"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-01", "family": "CURR", "rep": 0, "latency_s": 7.46, "context_chars": 5242, "prompt_chars": 6157, "raw": "```json\n{\"answer\": \"Kestrel Data Works\", \"source_id\": \"S-DOC-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Joined 2025-03-01, confirmed still employed as of 2025-02-21\"}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Joined 2025-03-01, confirmed still employed as of 2025-02-21"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "ENT-05", "family": "ENT", "rep": 0, "latency_s": 20.89, "context_chars": 6744, "prompt_chars": 7691, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner listed as colleague at Kestrel Data Works\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner listed as colleague at Kestrel Data Works"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HOP-06", "family": "HOP", "rep": 0, "latency_s": 9.57, "context_chars": 10038, "prompt_chars": 10987, "raw": "```json\n{\n  \"answer\": \"Tessera Analytics\",\n  \"source_id\": \"F-011\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Aurora 14 acquired 2024-02-20; Tessera Analytics employment valid through 2025-02-28.\"\n}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "F-011", "policy_id": null, "unanswerable": false, "stale": false, "note": "Aurora 14 acquired 2024-02-20; Tessera Analytics employment valid through 2025-02-28."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CONF-06", "family": "CONF", "rep": 1, "latency_s": 21.87, "context_chars": 5242, "prompt_chars": 6208, "raw": "```json\n{\n  \"answer\": \"The payslip (S-DOC-09) showing EUR 4410 should be believed; official employer documents are more reliable than personal notes.\",\n  \"source_id\": \"S-DOC-09\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Official payslip more reliable than personal note\"\n}\n```\n", "obj": {"answer": "The payslip (S-DOC-09) showing EUR 4410 should be believed; official employer documents are more reliable than personal notes.", "source_id": "S-DOC-09", "policy_id": null, "unanswerable": false, "stale": false, "note": "Official payslip more reliable than personal note"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "STALE-05", "family": "STALE", "rep": 0, "latency_s": 7.45, "context_chars": 4664, "prompt_chars": 5592, "raw": "```json\n{\n  \"answer\": \"Yes, until 2027-02-20\",\n  \"source_id\": \"S-EML-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Aurora 14 laptop warranty runs to 2027-02-20\"\n}\n```\n", "obj": {"answer": "Yes, until 2027-02-20", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Aurora 14 laptop warranty runs to 2027-02-20"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dimension", "item": "HIST-06", "family": "HIST", "rep": 0, "latency_s": 9.08, "context_chars": 10038, "prompt_chars": 10957, "raw": "```json\n{\n  \"answer\": \"vegetarian\",\n  \"source_id\": \"S-SLF-03\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Valid from 2023-06-02 to 2025-03-29\"\n}\n```\n", "obj": {"answer": "vegetarian", "source_id": "S-SLF-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2023-06-02 to 2025-03-29"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-07", "family": "CURR", "rep": 0, "latency_s": 16.26, "context_chars": 10038, "prompt_chars": 10954, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Penicillin allergy documented since 2015\"}\n```\n", "obj": {"answer": "penicillin", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Penicillin allergy documented since 2015"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "STALE-01", "family": "STALE", "rep": 1, "latency_s": 9.28, "context_chars": 6744, "prompt_chars": 7721, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Passport number not found in records\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Passport number not found in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "ENT-05", "family": "ENT", "rep": 1, "latency_s": 7.59, "context_chars": 2097, "prompt_chars": 3044, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner listed as colleague at Kestrel Data Works\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner listed as colleague at Kestrel Data Works"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-02", "family": "CONF", "rep": 1, "latency_s": 48.98, "context_chars": 4664, "prompt_chars": 5594, "raw": "```json\n{\n  \"answer\": \"EUR 4410\",\n  \"source_id\": \"S-DOC-09\",\n  \"policy_id\": \"official_precedence\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Payslip shows EUR 4410 from 2026-01-01. Personal note conflicts with EUR 4200.\"\n}\n```\n", "obj": {"answer": "EUR 4410", "source_id": "S-DOC-09", "policy_id": "official_precedence", "unanswerable": false, "stale": false, "note": "Payslip shows EUR 4410 from 2026-01-01. Personal note conflicts with EUR 4200."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CURR-07", "family": "CURR", "rep": 1, "latency_s": 7.81, "context_chars": 6744, "prompt_chars": 7660, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": \"F-064\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Drug allergy documented in medical record from 2024-11-09\"}\n```\n", "obj": {"answer": "penicillin", "source_id": "F-064", "policy_id": null, "unanswerable": false, "stale": false, "note": "Drug allergy documented in medical record from 2024-11-09"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HOP-01", "family": "HOP", "rep": 1, "latency_s": 14.18, "context_chars": 5242, "prompt_chars": 6200, "raw": "```json\n{\n  \"answer\": \"Nadja Lind\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Most recent Nicosia trip was 2026-01-22 with Nadja Lind. Sister Ilona lives in Nicosia.\"\n}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent Nicosia trip was 2026-01-22 with Nadja Lind. Sister Ilona lives in Nicosia."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HOP-02", "family": "HOP", "rep": 0, "latency_s": 8.71, "context_chars": 2097, "prompt_chars": 3048, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records do not contain employment history or residence information for the Tessera Analytics period.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records do not contain employment history or residence information for the Tessera Analytics period."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "STALE-02", "family": "STALE", "rep": 0, "latency_s": 23.59, "context_chars": 4664, "prompt_chars": 5587, "raw": "```json\n{\n  \"answer\": \"No\",\n  \"source_id\": \"S-DOC-08\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Policy expired 2026-08-31. Renewal status unknown.\"\n}\n```\n", "obj": {"answer": "No", "source_id": "S-DOC-08", "policy_id": null, "unanswerable": false, "stale": true, "note": "Policy expired 2026-08-31. Renewal status unknown."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dimension", "item": "HIST-04", "family": "HIST", "rep": 0, "latency_s": 10.83, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\"answer\": \"Halvard Systems\", \"source_id\": \"S-EML-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed there from 2018-09-01 to 2021-11-30.\"}\n```\n", "obj": {"answer": "Halvard Systems", "source_id": "S-EML-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed there from 2018-09-01 to 2021-11-30."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "STALE-04", "family": "STALE", "rep": 1, "latency_s": 9.96, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": \"F-061\",\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Licence number DL-772104 found; expiration date not recorded.\"\n}\n```\n", "obj": {"answer": null, "source_id": "F-061", "policy_id": null, "unanswerable": true, "stale": false, "note": "Licence number DL-772104 found; expiration date not recorded."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "STALE-03", "family": "STALE", "rep": 1, "latency_s": 12.08, "context_chars": 5242, "prompt_chars": 6171, "raw": "```json\n{\"answer\": \"No, the handset warranty expired on 2026-07-01.\", \"source_id\": \"S-EML-03\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Corvus 8 warranty ended 2026-07-01; today is 2026-09-06.\"}\n```\n", "obj": {"answer": "No, the handset warranty expired on 2026-07-01.", "source_id": "S-EML-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Corvus 8 warranty ended 2026-07-01; today is 2026-09-06."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "CONF-03", "family": "CONF", "rep": 0, "latency_s": 42.84, "context_chars": 5242, "prompt_chars": 6186, "raw": "```json\n{\n  \"answer\": \"12\",\n  \"source_id\": \"S-SLF-05\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Current rate since 2021-05-01; supersedes earlier billing record.\"\n}\n```\n", "obj": {"answer": "12", "source_id": "S-SLF-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Current rate since 2021-05-01; supersedes earlier billing record."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 1, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "GAP-02", "family": "GAP", "rep": 1, "latency_s": 7.75, "context_chars": 2097, "prompt_chars": 3010, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain no information about 2016 residence.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain no information about 2016 residence."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "CURR-03", "family": "CURR", "rep": 1, "latency_s": 6.24, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Driving licence from records\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Driving licence from records"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HIST-03", "family": "HIST", "rep": 1, "latency_s": 8.2, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records lack historical employment dates\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records lack historical employment dates"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "CURR-05", "family": "CURR", "rep": 0, "latency_s": 8.1, "context_chars": 4664, "prompt_chars": 5574, "raw": "```json\n{\n  \"answer\": \"Aurora 14\",\n  \"source_id\": \"S-EML-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Purchased 2024-02-20, warranty valid through 2027-02-20.\"\n}\n```\n", "obj": {"answer": "Aurora 14", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Purchased 2024-02-20, warranty valid through 2027-02-20."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HIST-01", "family": "HIST", "rep": 1, "latency_s": 9.94, "context_chars": 10038, "prompt_chars": 10957, "raw": "```json\n{\"answer\": \"48 Aristotelous, Nicosia\", \"source_id\": \"S-DOC-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2022-08-01 through 2025-05-14, covering 2023-01-01.\"}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2022-08-01 through 2025-05-14, covering 2023-01-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "ENT-01", "family": "ENT", "rep": 0, "latency_s": 13.33, "context_chars": 6744, "prompt_chars": 7688, "raw": "```json\n{\n  \"answer\": \"+357 99 018 442\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Kovacs is the landlord of 7 Kalymnou in Larnaca.\"\n}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs is the landlord of 7 Kalymnou in Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "STALE-02", "family": "STALE", "rep": 1, "latency_s": 24.26, "context_chars": 4664, "prompt_chars": 5587, "raw": "```json\n{\n  \"answer\": \"No, it expired on 2026-08-31.\",\n  \"source_id\": \"S-DOC-08\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Policy HI-88231 covered until 2026-08-31; today is 2026-09-06.\"\n}\n```\n", "obj": {"answer": "No, it expired on 2026-08-31.", "source_id": "S-DOC-08", "policy_id": null, "unanswerable": false, "stale": false, "note": "Policy HI-88231 covered until 2026-08-31; today is 2026-09-06."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "CURR-04", "family": "CURR", "rep": 1, "latency_s": 6.19, "context_chars": 4903, "prompt_chars": 5847, "raw": "```json\n{\"answer\": \"22\", \"source_id\": \"F-084\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Journal subscription costs 22 EUR monthly, valid since 2024-03-15\"}\n```\n", "obj": {"answer": "22", "source_id": "F-084", "policy_id": null, "unanswerable": false, "stale": false, "note": "Journal subscription costs 22 EUR monthly, valid since 2024-03-15"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-04", "family": "CONF", "rep": 0, "latency_s": 42.41, "context_chars": 4664, "prompt_chars": 5585, "raw": "```json\n{\"answer\": \"+357 96 442 018\", \"source_id\": \"S-DEV-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current from 2024-07-01; device-synced\"}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Current from 2024-07-01; device-synced"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-04", "family": "CURR", "rep": 0, "latency_s": 12.35, "context_chars": 4664, "prompt_chars": 5608, "raw": "```json\n{\"answer\": \"EUR 22\", \"source_id\": \"S-EML-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Journal access costs EUR 22 monthly since 2024-03-15.\"}\n```\n", "obj": {"answer": "EUR 22", "source_id": "S-EML-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Journal access costs EUR 22 monthly since 2024-03-15."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CONF-04", "family": "CONF", "rep": 1, "latency_s": 30.84, "context_chars": 4903, "prompt_chars": 5824, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018, +357 99 100 224\",\n  \"source_id\": \"F-021, F-022\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Two mobile numbers currently valid in records\"\n}\n```\n", "obj": {"answer": "+357 96 442 018, +357 99 100 224", "source_id": "F-021, F-022", "policy_id": null, "unanswerable": false, "stale": false, "note": "Two mobile numbers currently valid in records"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CONF-02", "family": "CONF", "rep": 1, "latency_s": 28.45, "context_chars": 4903, "prompt_chars": 5833, "raw": "```json\n{\"answer\": \"4200\", \"source_id\": \"F-032\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Two valid records for 2026; chose F-032 over conflicting F-031\"}\n```\n", "obj": {"answer": "4200", "source_id": "F-032", "policy_id": null, "unanswerable": false, "stale": false, "note": "Two valid records for 2026; chose F-032 over conflicting F-031"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-04", "family": "HIST", "rep": 0, "latency_s": 8.2, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Only employer record (Kestrel Data Works) dates to 2025-02-21, after queried date.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Only employer record (Kestrel Data Works) dates to 2025-02-21, after queried date."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-07", "family": "HIST", "rep": 1, "latency_s": 8.72, "context_chars": 5242, "prompt_chars": 6197, "raw": "```json\n{\n  \"answer\": \"35\",\n  \"source_id\": \"S-EML-06\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Nicosia gym cost EUR 35/month from 2022-09-01 to 2025-05-31\"\n}\n```\n", "obj": {"answer": "35", "source_id": "S-EML-06", "policy_id": null, "unanswerable": false, "stale": false, "note": "Nicosia gym cost EUR 35/month from 2022-09-01 to 2025-05-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-06", "family": "CURR", "rep": 1, "latency_s": 9.1, "context_chars": 4664, "prompt_chars": 5573, "raw": "```json\n{\"answer\": \"CY17 0020 0128 0000 0012 0052 7600\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Bank account for salary; on file since 2021-12-01.\"}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Bank account for salary; on file since 2021-12-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HIST-07", "family": "HIST", "rep": 1, "latency_s": 9.51, "context_chars": 2097, "prompt_chars": 3052, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No gym payment information found in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No gym payment information found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "CONF-05", "family": "CONF", "rep": 1, "latency_s": 7.67, "context_chars": 2097, "prompt_chars": 3007, "raw": "```json\n{\"answer\": \"0+\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Blood type found in personal records.\"}\n```\n", "obj": {"answer": "0+", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Blood type found in personal records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HOP-06", "family": "HOP", "rep": 1, "latency_s": 51.84, "context_chars": 6744, "prompt_chars": 7693, "raw": "```json\n{\n  \"answer\": \"Kestrel Data Works\",\n  \"source_id\": \"S-DOC-05\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Laptop purchased 2024-02-20; employer documented 2025-02-21, no employment changes recorded.\"\n}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "S-DOC-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Laptop purchased 2024-02-20; employer documented 2025-02-21, no employment changes recorded."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "GAP-04", "family": "GAP", "rep": 1, "latency_s": 7.04, "context_chars": 6744, "prompt_chars": 7666, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister's email address not in records\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister's email address not in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HIST-07", "family": "HIST", "rep": 0, "latency_s": 7.82, "context_chars": 4664, "prompt_chars": 5619, "raw": "```json\n{\"answer\": \"EUR 35\", \"source_id\": \"S-EML-06\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Nicosia gym cost EUR 35/month from 2022-09-01 to 2025-05-31\"}\n```\n", "obj": {"answer": "EUR 35", "source_id": "S-EML-06", "policy_id": null, "unanswerable": false, "stale": false, "note": "Nicosia gym cost EUR 35/month from 2022-09-01 to 2025-05-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HOP-01", "family": "HOP", "rep": 0, "latency_s": 10.86, "context_chars": 10038, "prompt_chars": 10996, "raw": "```json\n{\n  \"answer\": \"Nadja Lind\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Sister Ilona lives in Nicosia. Most recent Nicosia trip: 2026-01-22 with Nadja Lind.\"\n}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Sister Ilona lives in Nicosia. Most recent Nicosia trip: 2026-01-22 with Nadja Lind."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "ENT-03", "family": "ENT", "rep": 0, "latency_s": 8.79, "context_chars": 2097, "prompt_chars": 3049, "raw": "```json\n{\"answer\": \"+357 96 220 145\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Berlin trip 2026-03-14 companion was Alex Reiner, colleague at Kestrel Data Works\"}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Berlin trip 2026-03-14 companion was Alex Reiner, colleague at Kestrel Data Works"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "GAP-05", "family": "GAP", "rep": 0, "latency_s": 8.86, "context_chars": 6744, "prompt_chars": 7683, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Employer identified as Kestrel Data Works but office street address not recorded.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Employer identified as Kestrel Data Works but office street address not recorded."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "STALE-04", "family": "STALE", "rep": 1, "latency_s": 6.22, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\"answer\": \"Yes, your driving licence DL-772104 is valid until 2027-01-31\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Licence valid until 2027-01-31\"}\n```\n", "obj": {"answer": "Yes, your driving licence DL-772104 is valid until 2027-01-31", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Licence valid until 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "flat", "item": "HIST-05", "family": "HIST", "rep": 0, "latency_s": 8.19, "context_chars": 2097, "prompt_chars": 3025, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records show current mobile number but lack historical data for 2023-03-01\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show current mobile number but lack historical data for 2023-03-01"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "STALE-01", "family": "STALE", "rep": 1, "latency_s": 10.76, "context_chars": 10038, "prompt_chars": 11015, "raw": "```json\n{\"answer\": \"P4417832. You cannot use it; it expired on 2026-04-11.\", \"source_id\": \"S-DOC-06\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"Passport expired; valid until 2026-04-11 only.\"}\n```\n", "obj": {"answer": "P4417832. You cannot use it; it expired on 2026-04-11.", "source_id": "S-DOC-06", "policy_id": null, "unanswerable": false, "stale": true, "note": "Passport expired; valid until 2026-04-11 only."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dimension", "item": "HIST-03", "family": "HIST", "rep": 1, "latency_s": 8.0, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\"answer\": \"Tessera Analytics\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed at Tessera Analytics from 2021-12-01 to 2025-02-28\"}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Tessera Analytics from 2021-12-01 to 2025-02-28"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "GAP-04", "family": "GAP", "rep": 0, "latency_s": 7.22, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister's email address is not in the records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister's email address is not in the records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HOP-04", "family": "HOP", "rep": 1, "latency_s": 48.64, "context_chars": 6744, "prompt_chars": 7723, "raw": "```json\n{\n  \"answer\": \"Alex Reiner, Alex Doran, Alex Kovacs, Petros Sarris\",\n  \"source_id\": \"F-003\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Four contacts live in Larnaca, Mira's address before Nicosia\"\n}\n```\n", "obj": {"answer": "Alex Reiner, Alex Doran, Alex Kovacs, Petros Sarris", "source_id": "F-003", "policy_id": null, "unanswerable": false, "stale": false, "note": "Four contacts live in Larnaca, Mira's address before Nicosia"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "GAP-03", "family": "GAP", "rep": 0, "latency_s": 7.19, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Vienna trip (2024-09-17) is recorded but cost information is not in the records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Vienna trip (2024-09-17) is recorded but cost information is not in the records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HIST-02", "family": "HIST", "rep": 0, "latency_s": 8.45, "context_chars": 4903, "prompt_chars": 5822, "raw": "```json\n{\n  \"answer\": \"12 Harbour Lane, Limassol\",\n  \"source_id\": \"F-001\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Address valid from 2019-03-01 to 2022-07-31\"\n}\n```\n", "obj": {"answer": "12 Harbour Lane, Limassol", "source_id": "F-001", "policy_id": null, "unanswerable": false, "stale": false, "note": "Address valid from 2019-03-01 to 2022-07-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "ENT-05", "family": "ENT", "rep": 1, "latency_s": 11.05, "context_chars": 4903, "prompt_chars": 5850, "raw": "```json\n{\n  \"answer\": \"Alex Reiner\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Currently employed at Kestrel Data Works since 2025-03-01\"\n}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Currently employed at Kestrel Data Works since 2025-03-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "ENT-04", "family": "ENT", "rep": 1, "latency_s": 8.49, "context_chars": 4664, "prompt_chars": 5599, "raw": "```json\n{\"answer\": \"3\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, Alex Doran, and Alex Kovacs in contacts.\"}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, and Alex Kovacs in contacts."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "ENT-02", "family": "ENT", "rep": 1, "latency_s": 9.16, "context_chars": 4903, "prompt_chars": 5834, "raw": "```json\n{\"answer\": \"Alex Doran, in Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran listed as general practitioner in Larnaca.\"}\n```\n", "obj": {"answer": "Alex Doran, in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran listed as general practitioner in Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CONF-03", "family": "CONF", "rep": 1, "latency_s": 27.29, "context_chars": 4903, "prompt_chars": 5847, "raw": "```json\n{\n  \"answer\": \"12\",\n  \"source_id\": \"F-081\",\n  \"policy_id\": \"conflict_resolution_higher_value\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Two Cloudline subscription records conflict; using higher value from F-081.\"\n}\n```\n", "obj": {"answer": "12", "source_id": "F-081", "policy_id": "conflict_resolution_higher_value", "unanswerable": false, "stale": false, "note": "Two Cloudline subscription records conflict; using higher value from F-081."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 1, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "GAP-01", "family": "GAP", "rep": 0, "latency_s": 7.94, "context_chars": 4664, "prompt_chars": 5589, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No national insurance number found in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No national insurance number found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HOP-02", "family": "HOP", "rep": 0, "latency_s": 10.06, "context_chars": 6744, "prompt_chars": 7695, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records mention Tessera Analytics as former employer of colleague Nadja Lind, but contain no start date or timeline for Mira's employment there.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records mention Tessera Analytics as former employer of colleague Nadja Lind, but contain no start date or timeline for Mira's employment there."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HOP-05", "family": "HOP", "rep": 1, "latency_s": 79.86, "context_chars": 6744, "prompt_chars": 7700, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Subscription values found but currency not explicitly specified.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Subscription values found but currency not explicitly specified."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "STALE-02", "family": "STALE", "rep": 0, "latency_s": 10.1, "context_chars": 5242, "prompt_chars": 6165, "raw": "```json\n{\"answer\": \"No. Your health insurance policy HI-88231 expired on 2026-08-31, before today (2026-09-06).\", \"source_id\": \"S-DOC-08\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Insurance expired 6 days ago. Renewal was due by 2026-08-31.\"}\n```\n", "obj": {"answer": "No. Your health insurance policy HI-88231 expired on 2026-08-31, before today (2026-09-06).", "source_id": "S-DOC-08", "policy_id": null, "unanswerable": false, "stale": false, "note": "Insurance expired 6 days ago. Renewal was due by 2026-08-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "HIST-05", "family": "HIST", "rep": 1, "latency_s": 9.7, "context_chars": 5242, "prompt_chars": 6170, "raw": "```json\n{\"answer\": \"+357 99 100 224\", \"source_id\": \"S-SLF-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2019-01-10 until 2024-06-30\"}\n```\n", "obj": {"answer": "+357 99 100 224", "source_id": "S-SLF-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2019-01-10 until 2024-06-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HIST-04", "family": "HIST", "rep": 1, "latency_s": 9.54, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\"answer\": \"Halvard Systems\", \"source_id\": \"S-EML-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employed 2018-09-01 to 2021-11-30\"}\n```\n", "obj": {"answer": "Halvard Systems", "source_id": "S-EML-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed 2018-09-01 to 2021-11-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "STALE-02", "family": "STALE", "rep": 0, "latency_s": 23.91, "context_chars": 6744, "prompt_chars": 7667, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": \"F-091\",\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": true,\n  \"note\": \"Health insurance review deadline 2026-08-31 has passed; current validity status unknown.\"\n}\n```\n", "obj": {"answer": null, "source_id": "F-091", "policy_id": null, "unanswerable": true, "stale": true, "note": "Health insurance review deadline 2026-08-31 has passed; current validity status unknown."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "ENT-01", "family": "ENT", "rep": 0, "latency_s": 11.14, "context_chars": 4903, "prompt_chars": 5847, "raw": "```json\n{\n  \"answer\": \"+357 99 018 442\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Kovacs, landlord of 7 Kalymnou, Larnaca (current address).\"\n}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs, landlord of 7 Kalymnou, Larnaca (current address)."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HIST-02", "family": "HIST", "rep": 1, "latency_s": 9.31, "context_chars": 10038, "prompt_chars": 10957, "raw": "```json\n{\"answer\": \"12 Harbour Lane, Limassol\", \"source_id\": \"S-DOC-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid 2019-03-01 to 2022-07-31\"}\n```\n", "obj": {"answer": "12 Harbour Lane, Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid 2019-03-01 to 2022-07-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CONF-02", "family": "CONF", "rep": 1, "latency_s": 44.16, "context_chars": 5242, "prompt_chars": 6172, "raw": "```json\n{\n  \"answer\": \"4410\",\n  \"source_id\": \"S-DOC-09\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Payslip shows EUR 4410 from Jan 2026; contradicted by personal note EUR 4200\"\n}\n```\n", "obj": {"answer": "4410", "source_id": "S-DOC-09", "policy_id": null, "unanswerable": false, "stale": false, "note": "Payslip shows EUR 4410 from Jan 2026; contradicted by personal note EUR 4200"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "ENT-04", "family": "ENT", "rep": 1, "latency_s": 9.86, "context_chars": 5242, "prompt_chars": 6177, "raw": "```json\n{\"answer\": \"3\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, Alex Doran, Alex Kovacs\"}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, Alex Kovacs"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HIST-01", "family": "HIST", "rep": 1, "latency_s": 8.22, "context_chars": 2097, "prompt_chars": 3016, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Records show current address but contain no historical address data for 2023-01-01.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show current address but contain no historical address data for 2023-01-01."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "GAP-04", "family": "GAP", "rep": 1, "latency_s": 6.36, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister Ilona Vantaa's phone is listed, but no email address in records.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister Ilona Vantaa's phone is listed, but no email address in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "ENT-01", "family": "ENT", "rep": 1, "latency_s": 10.43, "context_chars": 4664, "prompt_chars": 5608, "raw": "```json\n{\"answer\": \"+357 99 018 442\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Kovacs is landlord of 7 Kalymnou, Larnaca\"}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs is landlord of 7 Kalymnou, Larnaca"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HOP-03", "family": "HOP", "rep": 1, "latency_s": 25.62, "context_chars": 6744, "prompt_chars": 7691, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"F-021\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Device sync fact (tier 2) from 2024-07-01, most recent before Vienna trip 2024-09-17\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "F-021", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Device sync fact (tier 2) from 2024-07-01, most recent before Vienna trip 2024-09-17"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "GAP-01", "family": "GAP", "rep": 0, "latency_s": 7.8, "context_chars": 4903, "prompt_chars": 5828, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No national insurance number in records\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No national insurance number in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HIST-06", "family": "HIST", "rep": 1, "latency_s": 8.56, "context_chars": 6744, "prompt_chars": 7663, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Diet only recorded as of 2025-03-30; no data for 2024-01-01.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Diet only recorded as of 2025-03-30; no data for 2024-01-01."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "ENT-02", "family": "ENT", "rep": 1, "latency_s": 8.58, "context_chars": 5242, "prompt_chars": 6173, "raw": "```json\n{\"answer\": \"Alex Doran in Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran listed as GP in Larnaca.\"}\n```\n", "obj": {"answer": "Alex Doran in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran listed as GP in Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "ENT-01", "family": "ENT", "rep": 1, "latency_s": 7.48, "context_chars": 5242, "prompt_chars": 6186, "raw": "```json\n{\"answer\": \"+357 99 018 442\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Kovacs is the landlord of 7 Kalymnou, Larnaca.\"}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs is the landlord of 7 Kalymnou, Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HOP-04", "family": "HOP", "rep": 0, "latency_s": 30.33, "context_chars": 5242, "prompt_chars": 6221, "raw": "```json\n{\n  \"answer\": \"Ilona Vantaa and Nadja Lind\",\n  \"source_id\": \"S-DOC-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Nicosia was the address before current Larnaca residence.\"\n}\n```\n", "obj": {"answer": "Ilona Vantaa and Nadja Lind", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Nicosia was the address before current Larnaca residence."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HOP-05", "family": "HOP", "rep": 0, "latency_s": 64.09, "context_chars": 4664, "prompt_chars": 5620, "raw": "```json\n{\n  \"answer\": \"99\",\n  \"source_id\": \"S-SLF-02, S-SLF-05, S-EML-06, S-EML-05, S-EML-07\",\n  \"policy_id\": \"most_recent_record\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Five active subscriptions on 2024-06-01\"\n}\n```\n", "obj": {"answer": "99", "source_id": "S-SLF-02, S-SLF-05, S-EML-06, S-EML-05, S-EML-07", "policy_id": "most_recent_record", "unanswerable": false, "stale": false, "note": "Five active subscriptions on 2024-06-01"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HIST-02", "family": "HIST", "rep": 1, "latency_s": 8.87, "context_chars": 4664, "prompt_chars": 5583, "raw": "```json\n{\"answer\": \"12 Harbour Lane, Limassol\", \"source_id\": \"S-DOC-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Lived there from 2019-03-01 until 2022-07-31\"}\n```\n", "obj": {"answer": "12 Harbour Lane, Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Lived there from 2019-03-01 until 2022-07-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "STALE-04", "family": "STALE", "rep": 0, "latency_s": 7.89, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\"answer\": \"Yes, valid until 2027-01-31\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Driving licence DL-772104 expires 2027-01-31\"}\n```\n", "obj": {"answer": "Yes, valid until 2027-01-31", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Driving licence DL-772104 expires 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dimension", "item": "CONF-01", "family": "CONF", "rep": 0, "latency_s": 12.81, "context_chars": 10038, "prompt_chars": 10958, "raw": "```json\n{\"answer\": \"7 Kalymnou, Larnaca\", \"source_id\": \"F-003\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Conflicts with F-004; tier 1 source wins per CP-01.\"}\n```\n", "obj": {"answer": "7 Kalymnou, Larnaca", "source_id": "F-003", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Conflicts with F-004; tier 1 source wins per CP-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-05", "family": "CURR", "rep": 0, "latency_s": 7.26, "context_chars": 10038, "prompt_chars": 10948, "raw": "```json\n{\n  \"answer\": \"Aurora 14\",\n  \"source_id\": \"S-EML-02\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Device acquired 2024-02-20, warranty until 2027-02-20\"\n}\n```\n", "obj": {"answer": "Aurora 14", "source_id": "S-EML-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Device acquired 2024-02-20, warranty until 2027-02-20"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CONF-05", "family": "CONF", "rep": 0, "latency_s": 10.4, "context_chars": 10038, "prompt_chars": 10948, "raw": "```json\n{\n  \"answer\": \"0+\",\n  \"source_id\": \"S-DOC-10\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Document scan (tier 1) prevails over self-reported (tier 4) per conflict policy.\"\n}\n```\n", "obj": {"answer": "0+", "source_id": "S-DOC-10", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Document scan (tier 1) prevails over self-reported (tier 4) per conflict policy."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-08", "family": "CURR", "rep": 0, "latency_s": 10.33, "context_chars": 5242, "prompt_chars": 6189, "raw": "```json\n{\"answer\": \"11\", \"source_id\": \"S-SLF-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"EUR 11 per month since 2020-01-01\"}\n```\n", "obj": {"answer": "11", "source_id": "S-SLF-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "EUR 11 per month since 2020-01-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "ENT-03", "family": "ENT", "rep": 0, "latency_s": 11.92, "context_chars": 4664, "prompt_chars": 5616, "raw": "```json\n{\"answer\": \"+357 96 220 145\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, colleague at Kestrel Data Works, Larnaca.\"}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, colleague at Kestrel Data Works, Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "STALE-02", "family": "STALE", "rep": 0, "latency_s": 21.14, "context_chars": 2097, "prompt_chars": 3020, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": \"review_due_health_insurance\",\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Review due 2026-08-31 is overdue; validity status not documented\"\n}\n```\n", "obj": {"answer": null, "source_id": "review_due_health_insurance", "policy_id": null, "unanswerable": true, "stale": false, "note": "Review due 2026-08-31 is overdue; validity status not documented"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CONF-02", "family": "CONF", "rep": 0, "latency_s": 26.18, "context_chars": 6744, "prompt_chars": 7674, "raw": "```json\n{\"answer\": \"4410\", \"source_id\": \"F-031\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": true, \"note\": \"Document from Jan 2026 applied via tier conflict policy; salary data is 7 months old.\"}\n```\n", "obj": {"answer": "4410", "source_id": "F-031", "policy_id": "CP-01", "unanswerable": false, "stale": true, "note": "Document from Jan 2026 applied via tier conflict policy; salary data is 7 months old."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 1, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-01", "family": "CURR", "rep": 0, "latency_s": 7.76, "context_chars": 4903, "prompt_chars": 5818, "raw": "```json\n{\"answer\": \"Kestrel Data Works\", \"source_id\": \"F-012\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current employer as of 2026-09-06 with open end date.\"}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "F-012", "policy_id": null, "unanswerable": false, "stale": false, "note": "Current employer as of 2026-09-06 with open end date."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-03", "family": "CURR", "rep": 1, "latency_s": 6.2, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid until 2027-01-31\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "GAP-02", "family": "GAP", "rep": 1, "latency_s": 8.16, "context_chars": 5242, "prompt_chars": 6155, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No residential records found for 2016; earliest documented address is 2019.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No residential records found for 2016; earliest documented address is 2019."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "CURR-08", "family": "CURR", "rep": 1, "latency_s": 7.34, "context_chars": 10038, "prompt_chars": 10985, "raw": "```json\n{\"answer\": \"11 euro per month\", \"source_id\": \"F-085\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Music subscription ongoing since 2020-01-01, self-reported\"}\n```\n", "obj": {"answer": "11 euro per month", "source_id": "F-085", "policy_id": null, "unanswerable": false, "stale": false, "note": "Music subscription ongoing since 2020-01-01, self-reported"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-04", "family": "CURR", "rep": 1, "latency_s": 7.66, "context_chars": 5242, "prompt_chars": 6186, "raw": "```json\n{\"answer\": \"22\", \"source_id\": \"S-EML-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Journal access costs EUR 22/month since 2024-03-15\"}\n```\n", "obj": {"answer": "22", "source_id": "S-EML-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Journal access costs EUR 22/month since 2024-03-15"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HOP-01", "family": "HOP", "rep": 1, "latency_s": 9.71, "context_chars": 4903, "prompt_chars": 5861, "raw": "```json\n{\"answer\": \"Nadja Lind\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Most recent Nicosia trip was 2026-01-22 with Nadja Lind. Sister Ilona Vantaa lives in Nicosia.\"}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent Nicosia trip was 2026-01-22 with Nadja Lind. Sister Ilona Vantaa lives in Nicosia."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "GAP-04", "family": "GAP", "rep": 1, "latency_s": 6.93, "context_chars": 4664, "prompt_chars": 5586, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records contain sister's phone number but not email address.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records contain sister's phone number but not email address."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "ENT-02", "family": "ENT", "rep": 1, "latency_s": 8.19, "context_chars": 2097, "prompt_chars": 3028, "raw": "```json\n{\"answer\": \"Alex Doran, in Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran listed as general practitioner in people records.\"}\n```\n", "obj": {"answer": "Alex Doran, in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran listed as general practitioner in people records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-06", "family": "CONF", "rep": 0, "latency_s": 29.06, "context_chars": 4664, "prompt_chars": 5630, "raw": "```json\n{\n  \"answer\": \"EUR 4410\",\n  \"source_id\": \"S-DOC-09\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Official payslip document is more reliable than personal note.\"\n}\n```\n", "obj": {"answer": "EUR 4410", "source_id": "S-DOC-09", "policy_id": null, "unanswerable": false, "stale": false, "note": "Official payslip document is more reliable than personal note."}, "parsed": 1, "answer_correct": 0, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CURR-01", "family": "CURR", "rep": 1, "latency_s": 8.63, "context_chars": 4903, "prompt_chars": 5818, "raw": "```json\n{\"answer\": \"Kestrel Data Works\", \"source_id\": \"F-012\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Current employment began 2025-03-01\"}\n```\n", "obj": {"answer": "Kestrel Data Works", "source_id": "F-012", "policy_id": null, "unanswerable": false, "stale": false, "note": "Current employment began 2025-03-01"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "ENT-04", "family": "ENT", "rep": 1, "latency_s": 7.08, "context_chars": 4903, "prompt_chars": 5838, "raw": "```json\n{\n  \"answer\": \"3\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Reiner, Alex Doran, and Alex Kovacs listed in contacts.\"\n}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, and Alex Kovacs listed in contacts."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "STALE-01", "family": "STALE", "rep": 0, "latency_s": 8.94, "context_chars": 10038, "prompt_chars": 11015, "raw": "```json\n{\"answer\": \"Passport number: P4417832. It expired on 2026-04-11 and cannot be used for travel.\", \"source_id\": \"S-DOC-06\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"Passport expired 2026-04-11; renewal required for travel.\"}\n```\n", "obj": {"answer": "Passport number: P4417832. It expired on 2026-04-11 and cannot be used for travel.", "source_id": "S-DOC-06", "policy_id": null, "unanswerable": false, "stale": true, "note": "Passport expired 2026-04-11; renewal required for travel."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "flat", "item": "HOP-06", "family": "HOP", "rep": 1, "latency_s": 7.73, "context_chars": 2097, "prompt_chars": 3046, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Purchase date of Aurora 14 laptop is not recorded\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Purchase date of Aurora 14 laptop is not recorded"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HIST-04", "family": "HIST", "rep": 1, "latency_s": 8.4, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": \"Halvard Systems\", \"source_id\": \"F-010\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Employment period 2018-09-01 to 2021-11-30 covers the requested date.\"}\n```\n", "obj": {"answer": "Halvard Systems", "source_id": "F-010", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employment period 2018-09-01 to 2021-11-30 covers the requested date."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-03", "family": "CURR", "rep": 0, "latency_s": 9.39, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": \"DL-772104\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid until 2027-01-31 per driving licence scan S-DOC-07\"}\n```\n", "obj": {"answer": "DL-772104", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31 per driving licence scan S-DOC-07"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "HOP-04", "family": "HOP", "rep": 0, "latency_s": 12.18, "context_chars": 2097, "prompt_chars": 3076, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Records show current address in Nicosia but no explicit previous address.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show current address in Nicosia but no explicit previous address."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "HOP-01", "family": "HOP", "rep": 1, "latency_s": 11.67, "context_chars": 6744, "prompt_chars": 7702, "raw": "```json\n{\n  \"answer\": \"Nadja Lind\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Most recent Nicosia trip (2026-01-22) with Nadja Lind. Sister Ilona lives there.\"\n}\n```\n", "obj": {"answer": "Nadja Lind", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Most recent Nicosia trip (2026-01-22) with Nadja Lind. Sister Ilona lives there."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "GAP-03", "family": "GAP", "rep": 0, "latency_s": 7.86, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Vienna trip recorded (2024-09-17) but cost data not provided in records\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Vienna trip recorded (2024-09-17) but cost data not provided in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "CONF-02", "family": "CONF", "rep": 1, "latency_s": 12.51, "context_chars": 10038, "prompt_chars": 10968, "raw": "```json\n{\"answer\": \"4410 EUR\", \"source_id\": \"S-DOC-09\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Document scan (tier 1) takes precedence over self-reported (tier 4) for same date.\"}\n```\n", "obj": {"answer": "4410 EUR", "source_id": "S-DOC-09", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Document scan (tier 1) takes precedence over self-reported (tier 4) for same date."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HOP-06", "family": "HOP", "rep": 0, "latency_s": 10.33, "context_chars": 5242, "prompt_chars": 6191, "raw": "```json\n{\n  \"answer\": \"Tessera Analytics\",\n  \"source_id\": \"S-DOC-04\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Employed at Tessera Analytics from 2021-12-01 to 2025-02-28 when laptop was purchased 2024-02-20\"\n}\n```\n", "obj": {"answer": "Tessera Analytics", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "Employed at Tessera Analytics from 2021-12-01 to 2025-02-28 when laptop was purchased 2024-02-20"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "ENT-03", "family": "ENT", "rep": 0, "latency_s": 13.25, "context_chars": 5242, "prompt_chars": 6194, "raw": "```json\n{\"answer\": \"+357 96 220 145\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner travelled to Berlin on 2026-03-14\"}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner travelled to Berlin on 2026-03-14"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "STALE-03", "family": "STALE", "rep": 1, "latency_s": 9.64, "context_chars": 6744, "prompt_chars": 7673, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No handset warranty information recorded. Only laptop warranty on file.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No handset warranty information recorded. Only laptop warranty on file."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "CURR-06", "family": "CURR", "rep": 0, "latency_s": 9.32, "context_chars": 6744, "prompt_chars": 7653, "raw": "```json\n{\"answer\": \"CY17 0020 0128 0000 0012 0052 7600\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"From document scan captured 2021-11-18\"}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": false, "note": "From document scan captured 2021-11-18"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CONF-02", "family": "CONF", "rep": 0, "latency_s": 13.38, "context_chars": 10038, "prompt_chars": 10968, "raw": "```json\n{\n  \"answer\": \"4410\",\n  \"source_id\": \"S-DOC-09\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Document scan (tier 1) supersedes conflicting self-reported value of 4200 from Feb 2026.\"\n}\n```\n", "obj": {"answer": "4410", "source_id": "S-DOC-09", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Document scan (tier 1) supersedes conflicting self-reported value of 4200 from Feb 2026."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HOP-04", "family": "HOP", "rep": 0, "latency_s": 40.14, "context_chars": 10038, "prompt_chars": 11017, "raw": "```json\n{\n  \"answer\": \"Ilona Vantaa, Nadja Lind\",\n  \"source_id\": \"F-002\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Previous address Nicosia; current Larnaca\"\n}\n```\n", "obj": {"answer": "Ilona Vantaa, Nadja Lind", "source_id": "F-002", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Previous address Nicosia; current Larnaca"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "HIST-08", "family": "HIST", "rep": 1, "latency_s": 7.91, "context_chars": 10038, "prompt_chars": 10991, "raw": "```json\n{\"answer\": \"19 euro\", \"source_id\": \"S-EML-05\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Trainline subscription valid 2023-01-01 to 2024-12-31\"}\n```\n", "obj": {"answer": "19 euro", "source_id": "S-EML-05", "policy_id": null, "unanswerable": false, "stale": false, "note": "Trainline subscription valid 2023-01-01 to 2024-12-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-07", "family": "CURR", "rep": 1, "latency_s": 7.65, "context_chars": 2097, "prompt_chars": 3013, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Penicillin allergy recorded in health records\"}\n```\n", "obj": {"answer": "penicillin", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Penicillin allergy recorded in health records"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "STALE-01", "family": "STALE", "rep": 1, "latency_s": 16.43, "context_chars": 4903, "prompt_chars": 5880, "raw": "```json\n{\n  \"answer\": \"P4417832; expired 2026-04-11, cannot be used\",\n  \"source_id\": \"F-060\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Passport expired and is invalid for travel booking.\"\n}\n```\n", "obj": {"answer": "P4417832; expired 2026-04-11, cannot be used", "source_id": "F-060", "policy_id": null, "unanswerable": false, "stale": true, "note": "Passport expired and is invalid for travel booking."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "notes_profile", "item": "CURR-02", "family": "CURR", "rep": 0, "latency_s": 6.75, "context_chars": 5242, "prompt_chars": 6154, "raw": "```json\n{\"answer\": \"pescatarian\", \"source_id\": \"S-SLF-08\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Adopted pescatarian diet from 2025-03-30\"}\n```\n", "obj": {"answer": "pescatarian", "source_id": "S-SLF-08", "policy_id": null, "unanswerable": false, "stale": false, "note": "Adopted pescatarian diet from 2025-03-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HIST-05", "family": "HIST", "rep": 0, "latency_s": 10.69, "context_chars": 5242, "prompt_chars": 6170, "raw": "```json\n{\"answer\": \"+357 99 100 224\", \"source_id\": \"S-SLF-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Your number from 2019-01-10 until 2024-06-30\"}\n```\n", "obj": {"answer": "+357 99 100 224", "source_id": "S-SLF-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Your number from 2019-01-10 until 2024-06-30"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "GAP-02", "family": "GAP", "rep": 0, "latency_s": 8.5, "context_chars": 10038, "prompt_chars": 10951, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"No home address records available before 2019-03-01.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No home address records available before 2019-03-01."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "HIST-01", "family": "HIST", "rep": 0, "latency_s": 9.04, "context_chars": 2097, "prompt_chars": 3016, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records show current address at 48 Aristotelous, Nicosia, but no historical residence data for 2023-01-01.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records show current address at 48 Aristotelous, Nicosia, but no historical residence data for 2023-01-01."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HIST-02", "family": "HIST", "rep": 0, "latency_s": 8.6, "context_chars": 4664, "prompt_chars": 5583, "raw": "```json\n{\n  \"answer\": \"12 Harbour Lane, Limassol\",\n  \"source_id\": \"S-DOC-01\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Lived there from 2019-03-01 to 2022-07-31\"\n}\n```\n", "obj": {"answer": "12 Harbour Lane, Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Lived there from 2019-03-01 to 2022-07-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CONF-06", "family": "CONF", "rep": 1, "latency_s": 9.95, "context_chars": 10038, "prompt_chars": 11004, "raw": "```json\n{\"answer\": \"F-031 (4410 EUR): document_scan tier 1 source beats self_reported tier 4 source per CP-01.\", \"source_id\": \"S-DOC-09\", \"policy_id\": \"CP-01\", \"unanswerable\": false, \"stale\": false, \"note\": \"Both facts apply to 2026-09-06; lower tier wins.\"}\n```\n", "obj": {"answer": "F-031 (4410 EUR): document_scan tier 1 source beats self_reported tier 4 source per CP-01.", "source_id": "S-DOC-09", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Both facts apply to 2026-09-06; lower tier wins."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HOP-05", "family": "HOP", "rep": 0, "latency_s": 49.31, "context_chars": 4903, "prompt_chars": 5859, "raw": "```json\n{\n  \"answer\": \"108\",\n  \"source_id\": \"F-080,F-081,F-082,F-083,F-084,F-085\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Sum of six active subscriptions; two cloudline entries present\"\n}\n```\n", "obj": {"answer": "108", "source_id": "F-080,F-081,F-082,F-083,F-084,F-085", "policy_id": null, "unanswerable": false, "stale": false, "note": "Sum of six active subscriptions; two cloudline entries present"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "ENT-02", "family": "ENT", "rep": 0, "latency_s": 9.87, "context_chars": 4903, "prompt_chars": 5834, "raw": "```json\n{\"answer\": \"Alex Doran, Larnaca\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Doran is your general practitioner.\"}\n```\n", "obj": {"answer": "Alex Doran, Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran is your general practitioner."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HIST-02", "family": "HIST", "rep": 1, "latency_s": 8.92, "context_chars": 5242, "prompt_chars": 6161, "raw": "```json\n{\"answer\": \"12 Harbour Lane, Limassol\", \"source_id\": \"S-DOC-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Lease period 2019-03-01 to 2022-07-31 covers the date.\"}\n```\n", "obj": {"answer": "12 Harbour Lane, Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Lease period 2019-03-01 to 2022-07-31 covers the date."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HIST-07", "family": "HIST", "rep": 1, "latency_s": 8.15, "context_chars": 4664, "prompt_chars": 5619, "raw": "```json\n{\n  \"answer\": \"35\",\n  \"source_id\": \"S-EML-06\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Nicosia gym cost EUR 35/month from 2022-09-01 to 2025-05-31\"\n}\n```\n", "obj": {"answer": "35", "source_id": "S-EML-06", "policy_id": null, "unanswerable": false, "stale": false, "note": "Nicosia gym cost EUR 35/month from 2022-09-01 to 2025-05-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "STALE-04", "family": "STALE", "rep": 0, "latency_s": 8.14, "context_chars": 4903, "prompt_chars": 5825, "raw": "```json\n{\"answer\": \"Yes, valid until 2027-01-31\", \"source_id\": \"F-061\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Driving licence DL-772104 is current\"}\n```\n", "obj": {"answer": "Yes, valid until 2027-01-31", "source_id": "F-061", "policy_id": null, "unanswerable": false, "stale": false, "note": "Driving licence DL-772104 is current"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "dim_no_time", "item": "GAP-02", "family": "GAP", "rep": 0, "latency_s": 8.93, "context_chars": 6744, "prompt_chars": 7657, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Records document current addresses but lack residence dates for 2016.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records document current addresses but lack residence dates for 2016."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "STALE-03", "family": "STALE", "rep": 0, "latency_s": 7.19, "context_chars": 2097, "prompt_chars": 3026, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"No warranty expiration date recorded for phone\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No warranty expiration date recorded for phone"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HIST-08", "family": "HIST", "rep": 1, "latency_s": 9.11, "context_chars": 4903, "prompt_chars": 5856, "raw": "```json\n{\"answer\": \"19\", \"source_id\": \"F-082\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Trainline subscription was 19 EUR/month during July 2023\"}\n```\n", "obj": {"answer": "19", "source_id": "F-082", "policy_id": null, "unanswerable": false, "stale": false, "note": "Trainline subscription was 19 EUR/month during July 2023"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HOP-04", "family": "HOP", "rep": 0, "latency_s": 35.92, "context_chars": 6744, "prompt_chars": 7723, "raw": "```json\n{\n  \"answer\": \"Alex Reiner, Alex Doran, Alex Kovacs, Petros Sarris\",\n  \"source_id\": \"F-003\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Previous address Larnaca from F-003; current address Nicosia from F-004. Four contacts live in Larnaca.\"\n}\n```\n", "obj": {"answer": "Alex Reiner, Alex Doran, Alex Kovacs, Petros Sarris", "source_id": "F-003", "policy_id": null, "unanswerable": false, "stale": false, "note": "Previous address Larnaca from F-003; current address Nicosia from F-004. Four contacts live in Larnaca."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CURR-03", "family": "CURR", "rep": 0, "latency_s": 8.95, "context_chars": 10038, "prompt_chars": 10960, "raw": "```json\n{\n  \"answer\": \"DL-772104\",\n  \"source_id\": \"F-061\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Valid until 2027-01-31\"\n}\n```\n", "obj": {"answer": "DL-772104", "source_id": "F-061", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid until 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "GAP-04", "family": "GAP", "rep": 1, "latency_s": 6.57, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Sister identified but email address not in records\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Sister identified but email address not in records"}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "flat", "item": "CONF-06", "family": "CONF", "rep": 1, "latency_s": 11.49, "context_chars": 2097, "prompt_chars": 3063, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Only one salary record found (4200 EUR). No disagreement detected in provided records.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Only one salary record found (4200 EUR). No disagreement detected in provided records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dim_no_prov", "item": "HOP-05", "family": "HOP", "rep": 1, "latency_s": 98.57, "context_chars": 4903, "prompt_chars": 5859, "raw": "```json\n{\n  \"answer\": \"108\",\n  \"source_id\": \"F-080, F-081, F-082, F-083, F-084, F-085\",\n  \"policy_id\": \"Sum all active subscriptions including both cloudline records\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Resolved duplicate cloudline subscriptions by summing both active records during this period.\"\n}\n```\n", "obj": {"answer": "108", "source_id": "F-080, F-081, F-082, F-083, F-084, F-085", "policy_id": "Sum all active subscriptions including both cloudline records", "unanswerable": false, "stale": false, "note": "Resolved duplicate cloudline subscriptions by summing both active records during this period."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 1, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HIST-02", "family": "HIST", "rep": 1, "latency_s": 8.59, "context_chars": 4903, "prompt_chars": 5822, "raw": "```json\n{\"answer\": \"12 Harbour Lane, Limassol\", \"source_id\": \"F-001\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid address from 2019-03-01 to 2022-07-31\"}\n```\n", "obj": {"answer": "12 Harbour Lane, Limassol", "source_id": "F-001", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid address from 2019-03-01 to 2022-07-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "CONF-01", "family": "CONF", "rep": 0, "latency_s": 49.7, "context_chars": 4903, "prompt_chars": 5823, "raw": "```json\n{\n  \"answer\": \"7 Kalymnou, Larnaca and 48 Aristotelous, Nicosia\",\n  \"source_id\": \"F-003, F-004\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Two addresses both recorded as currently valid\"\n}\n```\n", "obj": {"answer": "7 Kalymnou, Larnaca and 48 Aristotelous, Nicosia", "source_id": "F-003, F-004", "policy_id": null, "unanswerable": false, "stale": false, "note": "Two addresses both recorded as currently valid"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "ENT-05", "family": "ENT", "rep": 1, "latency_s": 19.66, "context_chars": 10038, "prompt_chars": 10985, "raw": "```json\n{\"answer\": \"Alex Reiner\", \"source_id\": \"F-012\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner is listed as colleague at Kestrel Data Works, your current employer since 2025-03-01.\"}\n```\n", "obj": {"answer": "Alex Reiner", "source_id": "F-012", "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner is listed as colleague at Kestrel Data Works, your current employer since 2025-03-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "ENT-01", "family": "ENT", "rep": 0, "latency_s": 9.2, "context_chars": 4664, "prompt_chars": 5608, "raw": "```json\n{\n  \"answer\": \"+357 99 018 442\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Kovacs is the landlord of 7 Kalymnou in Larnaca.\"\n}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs is the landlord of 7 Kalymnou in Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-07", "family": "CURR", "rep": 0, "latency_s": 10.3, "context_chars": 4664, "prompt_chars": 5580, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Penicillin allergy recorded in laboratory file from 2024-11-09\"}\n```\n", "obj": {"answer": "penicillin", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Penicillin allergy recorded in laboratory file from 2024-11-09"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "ENT-04", "family": "ENT", "rep": 0, "latency_s": 7.5, "context_chars": 2097, "prompt_chars": 3032, "raw": "```json\n{\"answer\": \"3\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Reiner, Alex Doran, and Alex Kovacs in contacts\"}\n```\n", "obj": {"answer": "3", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner, Alex Doran, and Alex Kovacs in contacts"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CONF-06", "family": "CONF", "rep": 1, "latency_s": 17.21, "context_chars": 4664, "prompt_chars": 5630, "raw": "```json\n{\n  \"answer\": \"EUR 4410 from the payslip (S-DOC-09) should be believed\",\n  \"source_id\": \"S-DOC-09\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Payslip (S-DOC-09) shows EUR 4410 from 2026-01-01; own note (S-SLF-07) contradicts it with EUR 4200. Official document is authoritative.\"\n}\n```\n", "obj": {"answer": "EUR 4410 from the payslip (S-DOC-09) should be believed", "source_id": "S-DOC-09", "policy_id": null, "unanswerable": false, "stale": false, "note": "Payslip (S-DOC-09) shows EUR 4410 from 2026-01-01; own note (S-SLF-07) contradicts it with EUR 4200. Official document is authoritative."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "HIST-06", "family": "HIST", "rep": 1, "latency_s": 8.88, "context_chars": 5242, "prompt_chars": 6161, "raw": "```json\n{\n  \"answer\": \"vegetarian\",\n  \"source_id\": \"S-SLF-03\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Vegetarian diet held from 2023-06-02 until 2025-03-29.\"\n}\n```\n", "obj": {"answer": "vegetarian", "source_id": "S-SLF-03", "policy_id": null, "unanswerable": false, "stale": false, "note": "Vegetarian diet held from 2023-06-02 until 2025-03-29."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "HIST-08", "family": "HIST", "rep": 1, "latency_s": 12.14, "context_chars": 6744, "prompt_chars": 7697, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"No Trainline pass subscription found in records.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "No Trainline pass subscription found in records."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "CURR-04", "family": "CURR", "rep": 1, "latency_s": 9.77, "context_chars": 4664, "prompt_chars": 5608, "raw": "```json\n{\"answer\": \"EUR 22\", \"source_id\": \"S-EML-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Monthly cost since March 2024\"}\n```\n", "obj": {"answer": "EUR 22", "source_id": "S-EML-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Monthly cost since March 2024"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "ENT-03", "family": "ENT", "rep": 1, "latency_s": 12.63, "context_chars": 10038, "prompt_chars": 10990, "raw": "```json\n{\n  \"answer\": \"+357 96 220 145\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Reiner travelled to Berlin 2026-03-14; phone from people records.\"\n}\n```\n", "obj": {"answer": "+357 96 220 145", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Reiner travelled to Berlin 2026-03-14; phone from people records."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "ENT-01", "family": "ENT", "rep": 1, "latency_s": 8.11, "context_chars": 2097, "prompt_chars": 3041, "raw": "```json\n{\"answer\": \"+357 99 018 442\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Alex Kovacs, landlord of 7 Kalymnou in Larnaca\"}\n```\n", "obj": {"answer": "+357 99 018 442", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Kovacs, landlord of 7 Kalymnou in Larnaca"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "CURR-07", "family": "CURR", "rep": 0, "latency_s": 7.44, "context_chars": 5242, "prompt_chars": 6158, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Allergy confirmed by laboratory report 2024-11-09, known since 2015-01-01.\"}\n```\n", "obj": {"answer": "penicillin", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Allergy confirmed by laboratory report 2024-11-09, known since 2015-01-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_prov", "item": "HIST-06", "family": "HIST", "rep": 1, "latency_s": 6.78, "context_chars": 4903, "prompt_chars": 5822, "raw": "```json\n{\"answer\": \"vegetarian\", \"source_id\": \"F-050\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Valid from 2023-06-02 to 2025-03-29\"}\n```\n", "obj": {"answer": "vegetarian", "source_id": "F-050", "policy_id": null, "unanswerable": false, "stale": false, "note": "Valid from 2023-06-02 to 2025-03-29"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "flat", "item": "CURR-06", "family": "CURR", "rep": 1, "latency_s": 6.71, "context_chars": 2097, "prompt_chars": 3006, "raw": "```json\n{\"answer\": \"CY17 0020 0128 0000 0012 0052 7600\", \"source_id\": null, \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Bank IBAN from personal records\"}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Bank IBAN from personal records"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes_profile", "item": "STALE-04", "family": "STALE", "rep": 0, "latency_s": 13.24, "context_chars": 5242, "prompt_chars": 6164, "raw": "```json\n{\"answer\": \"Yes, it is valid until 2027-01-31\", \"source_id\": \"S-DOC-07\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Driving licence DL-772104 expires 2027-01-31\"}\n```\n", "obj": {"answer": "Yes, it is valid until 2027-01-31", "source_id": "S-DOC-07", "policy_id": null, "unanswerable": false, "stale": false, "note": "Driving licence DL-772104 expires 2027-01-31"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 1, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 1, "scored_false_stale": 0}
{"arm": "flat", "item": "HIST-04", "family": "HIST", "rep": 0, "latency_s": 7.16, "context_chars": 2097, "prompt_chars": 3019, "raw": "```json\n{\n  \"answer\": null,\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": true,\n  \"stale\": false,\n  \"note\": \"Records only show current employer, no historical dates.\"\n}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records only show current employer, no historical dates."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 1, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "dimension", "item": "CURR-07", "family": "CURR", "rep": 1, "latency_s": 8.47, "context_chars": 10038, "prompt_chars": 10954, "raw": "```json\n{\"answer\": \"penicillin\", \"source_id\": \"S-DOC-10\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Drug allergy recorded in medical document since 2015, validity open.\"}\n```\n", "obj": {"answer": "penicillin", "source_id": "S-DOC-10", "policy_id": null, "unanswerable": false, "stale": false, "note": "Drug allergy recorded in medical document since 2015, validity open."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "HOP-02", "family": "HOP", "rep": 0, "latency_s": 13.22, "context_chars": 4664, "prompt_chars": 5615, "raw": "```json\n{\"answer\": \"Limassol\", \"source_id\": \"S-DOC-01\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Tessera Analytics started 2021-12-01; Limassol lease ran to 2022-07-31.\"}\n```\n", "obj": {"answer": "Limassol", "source_id": "S-DOC-01", "policy_id": null, "unanswerable": false, "stale": false, "note": "Tessera Analytics started 2021-12-01; Limassol lease ran to 2022-07-31."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "GAP-05", "family": "GAP", "rep": 1, "latency_s": 7.27, "context_chars": 4664, "prompt_chars": 5603, "raw": "```json\n{\"answer\": null, \"source_id\": null, \"policy_id\": null, \"unanswerable\": true, \"stale\": false, \"note\": \"Records state employer based in Larnaca but no street address provided.\"}\n```\n", "obj": {"answer": null, "source_id": null, "policy_id": null, "unanswerable": true, "stale": false, "note": "Records state employer based in Larnaca but no street address provided."}, "parsed": 1, "answer_correct": 0, "source_correct": 0, "policy_correct": 0, "abstained": 1, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 0, "scored_source": 0, "scored_policy": 0, "scored_abst": 1, "scored_stale": 0, "scored_false_stale": 0}
{"arm": "notes", "item": "HIST-01", "family": "HIST", "rep": 1, "latency_s": 9.61, "context_chars": 4664, "prompt_chars": 5583, "raw": "```json\n{\"answer\": \"48 Aristotelous, Nicosia\", \"source_id\": \"S-DOC-02\", \"policy_id\": null, \"unanswerable\": false, \"stale\": false, \"note\": \"Lived there from 2022-08-01 until 2025-05-14\"}\n```\n", "obj": {"answer": "48 Aristotelous, Nicosia", "source_id": "S-DOC-02", "policy_id": null, "unanswerable": false, "stale": false, "note": "Lived there from 2022-08-01 until 2025-05-14"}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "ENT-02", "family": "ENT", "rep": 0, "latency_s": 11.61, "context_chars": 10038, "prompt_chars": 10969, "raw": "```json\n{\n  \"answer\": \"Alex Doran, in Larnaca\",\n  \"source_id\": null,\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Alex Doran is listed as general practitioner in Larnaca.\"\n}\n```\n", "obj": {"answer": "Alex Doran, in Larnaca", "source_id": null, "policy_id": null, "unanswerable": false, "stale": false, "note": "Alex Doran is listed as general practitioner in Larnaca."}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 0, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dimension", "item": "CONF-04", "family": "CONF", "rep": 0, "latency_s": 12.0, "context_chars": 10038, "prompt_chars": 10959, "raw": "```json\n{\n  \"answer\": \"+357 96 442 018\",\n  \"source_id\": \"S-DEV-01\",\n  \"policy_id\": \"CP-01\",\n  \"unanswerable\": false,\n  \"stale\": false,\n  \"note\": \"Device sync (tier 2) wins over conflicting self-reported (tier 4) from 2024-07-01.\"\n}\n```\n", "obj": {"answer": "+357 96 442 018", "source_id": "S-DEV-01", "policy_id": "CP-01", "unanswerable": false, "stale": false, "note": "Device sync (tier 2) wins over conflicting self-reported (tier 4) from 2024-07-01."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 1, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 0, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 1, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "notes", "item": "CURR-06", "family": "CURR", "rep": 0, "latency_s": 54.8, "context_chars": 4664, "prompt_chars": 5573, "raw": "```json\n{\"answer\": \"CY17 0020 0128 0000 0012 0052 7600\", \"source_id\": \"S-DOC-04\", \"policy_id\": null, \"unanswerable\": false, \"stale\": true, \"note\": \"From 2021; may need verification after 2025 employment change.\"}\n```\n", "obj": {"answer": "CY17 0020 0128 0000 0012 0052 7600", "source_id": "S-DOC-04", "policy_id": null, "unanswerable": false, "stale": true, "note": "From 2021; may need verification after 2025 employment change."}, "parsed": 1, "answer_correct": 1, "source_correct": 1, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 1, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}
{"arm": "dim_no_time", "item": "CURR-02", "family": "CURR", "rep": 1, "latency_s": 26.37, "context_chars": 6744, "prompt_chars": 7656, "raw": "```json\n{\n  \"answer\": \"pescatarian\",\n  \"source_id\": \"F-051\",\n  \"policy_id\": null,\n  \"unanswerable\": false,\n  \"stale\": true,\n  \"note\": \"Self-reported March 2025; over 1 year old for current query\"\n}\n```\n", "obj": {"answer": "pescatarian", "source_id": "F-051", "policy_id": null, "unanswerable": false, "stale": true, "note": "Self-reported March 2025; over 1 year old for current query"}, "parsed": 1, "answer_correct": 1, "source_correct": 0, "policy_correct": 0, "abstained": 0, "false_abstention": 0, "stale_correct": 0, "false_stale": 1, "overclaim": 0, "fabricated": 0, "scored_answer": 1, "scored_source": 1, "scored_policy": 0, "scored_abst": 0, "scored_stale": 0, "scored_false_stale": 1}