docs/benchmarks/context-activation-product-2026-09.json
Rama main · réplica de lectura del repositorio original.
{
"audit_passed": false,
"blocked": {
"count": 0,
"total": 18
},
"c9_padding_robustness": {
"base_case_id": "C2",
"padded_case_id": "C9",
"passed": false,
"precision_delta": null,
"precision_padded": null,
"reasons": [
"padded-tree precision below the 0.8 floor"
],
"recall_base": 0.5,
"recall_padded": 0.5
},
"fixture_mechanisms": {
"C1": "resolver",
"C2": "resolver",
"C3": "resolver",
"C4": "resolver",
"C5": "records_state",
"C6": "naming_gate",
"C7": "competing_senses",
"C8": "naming_gate",
"C9": "resolver",
"T1": "naming_gate",
"T2": "naming_gate",
"T3": "resolver",
"T4": "competing_senses",
"T5": "resolver",
"T6": "soundness",
"T7": "competing_senses",
"T8": "naming_gate",
"T9": "naming_gate"
},
"fixture_set_digest": "a49d85f49b18c2ce8f0349933ed01ceb4fb5ca176dca066700c9c2f93605426f",
"hedged_twins": {
"count": 0,
"total": 4
},
"instrument": "context-activation deterministic audit, real compiler",
"mechanism": "product_activate_context",
"mechanism_removal": {
"competing_senses": {
"audit_passed": false,
"passing_cases": [
"T1",
"T2",
"C5",
"T5",
"C6",
"T9"
],
"removes": "competing-sense abstention: disconnected same-kind resolved anchors never abstain"
},
"naming_gate": {
"audit_passed": false,
"passing_cases": [
"T1",
"T2",
"C5",
"T5",
"C6",
"C7",
"T9"
],
"removes": "the retrieval carry's naming gate: the top lexical hit for the turn's words is carried, with no corpus-rare phrase, corpus-size floor or single-page rule"
},
"records_state": {
"audit_passed": false,
"passing_cases": [
"T1",
"T2",
"T5",
"C6",
"C7",
"T9"
],
"removes": "governed current state: no Records or profile state reaches the packet"
},
"resolver": {
"audit_passed": false,
"passing_cases": [
"T1",
"T2",
"C6",
"T6",
"T9"
],
"removes": "anchor resolution: no candidate resolves, so no anchor, lane or current state is served"
},
"soundness": {
"audit_passed": false,
"passing_cases": [
"T1",
"T2",
"T4",
"C5",
"T5",
"C6",
"T9"
],
"removes": "the soundness rule: any one contact kind (a rare term or retrieval alone) resolves"
},
"working_set": {
"audit_passed": false,
"passing_cases": [
"T1",
"T2",
"C6",
"T6",
"T9"
],
"removes": "the whole compiler: the documented EXOMEM_DISABLE_WORKING_SET kill switch"
}
},
"per_case": [
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 2
},
"kind": "note",
"poison": {
"hit": 0,
"total": 1
}
},
{
"gold": {
"hit": 1,
"total": 1
},
"kind": "records_collection",
"poison": {
"hit": 0,
"total": 0
}
}
],
"case_id": "C1",
"expected_status": "resolved",
"failure_reasons": [
"gold recall 0.33 below the 0.9 floor",
"missing required fact(s): ['weekly limit', 'capacity ceiling']"
],
"gold": {
"hit": 1,
"total": 3
},
"hedged": false,
"is_twin": false,
"must_exclude_present": [],
"must_include_missing": [
"weekly limit",
"capacity ceiling"
],
"observed_status": "resolved",
"packet_chars": 325,
"passed": false,
"poison": {
"hit": 0,
"total": 1
},
"precision": 1.0,
"status_match": true,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "note",
"poison": {
"hit": 0,
"total": 2
}
},
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "records_collection",
"poison": {
"hit": 0,
"total": 1
}
}
],
"case_id": "T1",
"expected_status": "unresolved",
"failure_reasons": [],
"gold": {
"hit": 0,
"total": 0
},
"hedged": false,
"is_twin": true,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "unresolved",
"packet_chars": 0,
"passed": true,
"poison": {
"hit": 0,
"total": 3
},
"precision": null,
"status_match": true,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 1,
"total": 1
},
"kind": "equipment",
"poison": {
"hit": 0,
"total": 1
}
},
{
"gold": {
"hit": 0,
"total": 1
},
"kind": "note",
"poison": {
"hit": 0,
"total": 0
}
}
],
"case_id": "C2",
"expected_status": "resolved",
"failure_reasons": [
"expected status 'resolved', observed 'unresolved'",
"gold recall 0.50 below the 0.9 floor",
"missing required fact(s): ['grill', 'cooking method']"
],
"gold": {
"hit": 1,
"total": 2
},
"hedged": false,
"is_twin": false,
"must_exclude_present": [],
"must_include_missing": [
"grill",
"cooking method"
],
"observed_status": "unresolved",
"packet_chars": 20,
"passed": false,
"poison": {
"hit": 0,
"total": 1
},
"precision": null,
"status_match": false,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "equipment",
"poison": {
"hit": 0,
"total": 1
}
},
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "note",
"poison": {
"hit": 0,
"total": 1
}
}
],
"case_id": "T2",
"expected_status": "unresolved",
"failure_reasons": [],
"gold": {
"hit": 0,
"total": 0
},
"hedged": false,
"is_twin": true,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "unresolved",
"packet_chars": 0,
"passed": true,
"poison": {
"hit": 0,
"total": 2
},
"precision": null,
"status_match": true,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 1
},
"kind": "note",
"poison": {
"hit": 0,
"total": 0
}
},
{
"gold": {
"hit": 0,
"total": 1
},
"kind": "planning_item",
"poison": {
"hit": 0,
"total": 1
}
}
],
"case_id": "C3",
"expected_status": "resolved",
"failure_reasons": [
"expected status 'resolved', observed 'unresolved'",
"gold recall 0.00 below the 0.9 floor",
"missing required fact(s): ['extending the reporting module']"
],
"gold": {
"hit": 0,
"total": 2
},
"hedged": false,
"is_twin": false,
"must_exclude_present": [],
"must_include_missing": [
"extending the reporting module"
],
"observed_status": "unresolved",
"packet_chars": 46,
"passed": false,
"poison": {
"hit": 0,
"total": 1
},
"precision": null,
"status_match": false,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "note",
"poison": {
"hit": 0,
"total": 1
}
},
{
"gold": {
"hit": 0,
"total": 1
},
"kind": "planning_item",
"poison": {
"hit": 0,
"total": 1
}
}
],
"case_id": "T3",
"expected_status": "resolved",
"failure_reasons": [
"twin surfaced a ref outside its own gold",
"gold recall 0.00 below the 0.9 floor",
"precision 0.00 below the 0.8 floor"
],
"gold": {
"hit": 0,
"total": 1
},
"hedged": false,
"is_twin": true,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "resolved",
"packet_chars": 154,
"passed": false,
"poison": {
"hit": 0,
"total": 2
},
"precision": 0.0,
"status_match": true,
"twin_false_activation": true
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 1
},
"kind": "entity",
"poison": {
"hit": 0,
"total": 2
}
},
{
"gold": {
"hit": 0,
"total": 1
},
"kind": "note",
"poison": {
"hit": 0,
"total": 0
}
}
],
"case_id": "C4",
"expected_status": "resolved",
"failure_reasons": [
"expected status 'resolved', observed 'unresolved'",
"gold recall 0.00 below the 0.9 floor",
"missing required fact(s): ['failed on the Mac build']"
],
"gold": {
"hit": 0,
"total": 2
},
"hedged": false,
"is_twin": false,
"must_exclude_present": [],
"must_include_missing": [
"failed on the Mac build"
],
"observed_status": "unresolved",
"packet_chars": 0,
"passed": false,
"poison": {
"hit": 0,
"total": 2
},
"precision": null,
"status_match": false,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 2,
"total": 2
},
"kind": "entity",
"poison": {
"hit": 0,
"total": 1
}
},
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "note",
"poison": {
"hit": 0,
"total": 1
}
}
],
"case_id": "T4",
"expected_status": "ambiguous",
"failure_reasons": [
"expected status 'ambiguous', observed 'unresolved'"
],
"gold": {
"hit": 2,
"total": 2
},
"hedged": false,
"is_twin": true,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "unresolved",
"packet_chars": 21,
"passed": false,
"poison": {
"hit": 0,
"total": 2
},
"precision": null,
"status_match": false,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 1,
"total": 1
},
"kind": "records_collection",
"poison": {
"hit": 0,
"total": 0
}
},
{
"gold": {
"hit": 1,
"total": 1
},
"kind": "resource",
"poison": {
"hit": 0,
"total": 1
}
}
],
"case_id": "C5",
"expected_status": "resolved",
"failure_reasons": [],
"gold": {
"hit": 2,
"total": 2
},
"hedged": false,
"is_twin": false,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "resolved",
"packet_chars": 435,
"passed": true,
"poison": {
"hit": 0,
"total": 1
},
"precision": 1.0,
"status_match": true,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "records_collection",
"poison": {
"hit": 0,
"total": 1
}
},
{
"gold": {
"hit": 1,
"total": 1
},
"kind": "resource",
"poison": {
"hit": 0,
"total": 0
}
}
],
"case_id": "T5",
"expected_status": "resolved",
"failure_reasons": [],
"gold": {
"hit": 1,
"total": 1
},
"hedged": false,
"is_twin": true,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "resolved",
"packet_chars": 321,
"passed": true,
"poison": {
"hit": 0,
"total": 1
},
"precision": 1.0,
"status_match": true,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [],
"case_id": "C6",
"expected_status": "unresolved",
"failure_reasons": [],
"gold": {
"hit": 0,
"total": 0
},
"hedged": false,
"is_twin": false,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "unresolved",
"packet_chars": 0,
"passed": true,
"poison": {
"hit": 0,
"total": 0
},
"precision": null,
"status_match": true,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "equipment",
"poison": {
"hit": 1,
"total": 1
}
}
],
"case_id": "T6",
"expected_status": "unresolved",
"failure_reasons": [
"1 poison anchor(s) surfaced"
],
"gold": {
"hit": 0,
"total": 0
},
"hedged": false,
"is_twin": true,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "unresolved",
"packet_chars": 20,
"passed": false,
"poison": {
"hit": 1,
"total": 1
},
"precision": null,
"status_match": true,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 3,
"total": 3
},
"kind": "hub",
"poison": {
"hit": 0,
"total": 0
}
}
],
"case_id": "C7",
"expected_status": "ambiguous",
"failure_reasons": [],
"gold": {
"hit": 3,
"total": 3
},
"hedged": false,
"is_twin": false,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "ambiguous",
"packet_chars": 334,
"passed": true,
"poison": {
"hit": 0,
"total": 0
},
"precision": 1.0,
"status_match": true,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 1,
"total": 1
},
"kind": "hub",
"poison": {
"hit": 2,
"total": 2
}
}
],
"case_id": "T7",
"expected_status": "resolved",
"failure_reasons": [
"expected status 'resolved', observed 'ambiguous'",
"2 poison anchor(s) surfaced",
"twin surfaced a ref outside its own gold",
"precision 0.50 below the 0.8 floor"
],
"gold": {
"hit": 1,
"total": 1
},
"hedged": false,
"is_twin": true,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "ambiguous",
"packet_chars": 334,
"passed": false,
"poison": {
"hit": 2,
"total": 2
},
"precision": 0.5,
"status_match": false,
"twin_false_activation": true
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 1
},
"kind": "note",
"poison": {
"hit": 0,
"total": 2
}
}
],
"case_id": "C8",
"expected_status": "resolved",
"failure_reasons": [
"expected status 'resolved', observed 'unresolved'",
"gold recall 0.00 below the 0.9 floor",
"missing required fact(s): ['version 3']"
],
"gold": {
"hit": 0,
"total": 1
},
"hedged": false,
"is_twin": false,
"must_exclude_present": [],
"must_include_missing": [
"version 3"
],
"observed_status": "unresolved",
"packet_chars": 0,
"passed": false,
"poison": {
"hit": 0,
"total": 2
},
"precision": null,
"status_match": false,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 1
},
"kind": "note",
"poison": {
"hit": 0,
"total": 2
}
}
],
"case_id": "T8",
"expected_status": "resolved",
"failure_reasons": [
"expected status 'resolved', observed 'unresolved'",
"gold recall 0.00 below the 0.9 floor"
],
"gold": {
"hit": 0,
"total": 1
},
"hedged": false,
"is_twin": true,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "unresolved",
"packet_chars": 0,
"passed": false,
"poison": {
"hit": 0,
"total": 2
},
"precision": null,
"status_match": false,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 1,
"total": 1
},
"kind": "equipment",
"poison": {
"hit": 0,
"total": 1
}
},
{
"gold": {
"hit": 0,
"total": 1
},
"kind": "note",
"poison": {
"hit": 0,
"total": 0
}
}
],
"case_id": "C9",
"expected_status": "resolved",
"failure_reasons": [
"expected status 'resolved', observed 'unresolved'",
"gold recall 0.50 below the 0.9 floor",
"missing required fact(s): ['grill', 'cooking method']"
],
"gold": {
"hit": 1,
"total": 2
},
"hedged": false,
"is_twin": false,
"must_exclude_present": [],
"must_include_missing": [
"grill",
"cooking method"
],
"observed_status": "unresolved",
"packet_chars": 20,
"passed": false,
"poison": {
"hit": 0,
"total": 1
},
"precision": null,
"status_match": false,
"twin_false_activation": false
},
{
"blocked": false,
"by_anchor_kind": [
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "equipment",
"poison": {
"hit": 0,
"total": 1
}
},
{
"gold": {
"hit": 0,
"total": 0
},
"kind": "note",
"poison": {
"hit": 0,
"total": 1
}
}
],
"case_id": "T9",
"expected_status": "unresolved",
"failure_reasons": [],
"gold": {
"hit": 0,
"total": 0
},
"hedged": false,
"is_twin": true,
"must_exclude_present": [],
"must_include_missing": [],
"observed_status": "unresolved",
"packet_chars": 0,
"passed": true,
"poison": {
"hit": 0,
"total": 2
},
"precision": null,
"status_match": true,
"twin_false_activation": false
}
],
"threshold_digest": "7b2785cf81437ad98eeede3fb9bb046baa80fdcddf883d33c293cae91fbf867a",
"thresholds": {
"end_to_end_p50_ms_ceiling": 10744.641,
"end_to_end_p95_ms_ceiling": 16488.201,
"gold_recall_floor": 0.9,
"hedged_twins_ceiling": 1,
"padding_precision_floor": 0.8,
"precision_floor": 0.8,
"statement_max_chars": 200,
"token_hard_cap": 2000,
"token_p50_ceiling": 900,
"token_p95_ceiling": 1500,
"working_set_p50_ms_ceiling": 800,
"working_set_p95_ms_ceiling": 2500
},
"trees": [
{
"distractor_count": 0,
"fixtures": [
"C1",
"T1",
"C2",
"T2",
"C3",
"T3",
"C4",
"T4",
"C5",
"T5",
"C6",
"T6",
"C7",
"T7",
"C8",
"T8"
],
"graph": {
"edges": 45,
"indexed_files": 38,
"nodes": 65
},
"indexed_pages": 37,
"logical_corpus_digest": "0b8a8eef2bdf0d2461520f51b5e7a8e46a54ff61f417dfd6827e7097a284ffcb",
"published_anchors": 18,
"topology_findings": [
{
"case_id": "C1",
"finding": "no anchor, and the retrieval carry refuses a corpus of 37 indexed pages (minimum 100)",
"key": "c1_weekly_limit_insight",
"key_kind": "note"
},
{
"case_id": "C1",
"finding": "no anchor, and the retrieval carry refuses a corpus of 37 indexed pages (minimum 100)",
"key": "c1_capacity_ceilings_pattern",
"key_kind": "note"
},
{
"case_id": "C2",
"finding": "no anchor, and the retrieval carry refuses a corpus of 37 indexed pages (minimum 100)",
"key": "c2_cooking_method_insight",
"key_kind": "note"
},
{
"case_id": "C3",
"finding": "planning identity: the published plan anchor is Knowledge Base/Planning/Roadmap/_collection.md, not the gold item",
"key": "c3_planning_item",
"key_kind": "planning_item"
},
{
"case_id": "C3",
"finding": "no anchor, and the retrieval carry refuses a corpus of 37 indexed pages (minimum 100)",
"key": "c3_design_pointer",
"key_kind": "note"
},
{
"case_id": "T3",
"finding": "planning identity: the published plan anchor is Knowledge Base/Planning/Other Workstream/_collection.md, not the gold item",
"key": "t3_other_project_planning_item",
"key_kind": "planning_item"
},
{
"case_id": "C4",
"finding": "no anchor, and the retrieval carry refuses a corpus of 37 indexed pages (minimum 100)",
"key": "c4_failure_note",
"key_kind": "note"
},
{
"case_id": "C8",
"finding": "no anchor, and the retrieval carry refuses a corpus of 37 indexed pages (minimum 100)",
"key": "c8_active_head",
"key_kind": "note"
},
{
"case_id": "T8",
"finding": "no anchor, and the retrieval carry refuses a corpus of 37 indexed pages (minimum 100)",
"key": "t8_unchained_active_note",
"key_kind": "note"
}
]
},
{
"distractor_count": 200,
"fixtures": [
"C9",
"T9"
],
"graph": {
"edges": 45,
"indexed_files": 238,
"nodes": 265
},
"indexed_pages": 37,
"logical_corpus_digest": "3092af49a017e75c672992f6b9fbb16daa08dfcd0ab5eb58750de62cd0b5abf3",
"published_anchors": 18,
"topology_findings": [
{
"case_id": "C9",
"finding": "no anchor, and the retrieval carry refuses a corpus of 37 indexed pages (minimum 100)",
"key": "c2_cooking_method_insight",
"key_kind": "note"
}
]
}
],
"verdict": "reviewed"
}