Shay 92e9631467 docs: two lift attempts falsified + the ADR-0207 architectural impasse ... Built and measured two committing readers on held-out (not just diagnosed): - resolve_pooled composer: 2 correct / 87 wrong (17%) - maximally-narrow forced reader: 0 correct / 61 wrong (100% when it fires) Shallow committing CANNOT be sound on real GSM8K (measured). The impasse: - candidate-graph path is SOUND (wrong=0) but FROZEN by ADR-0207 §4, covers 0 real - composer is OPEN (ADR-0207 §5) but UNSOUND (17% wrong, no separating gate) ADR-0207's premise (composer is the wrong=0-safe path) is FALSIFIED by held-out data; 'feed the composer' is a path to more confabulation. Needs a follow-up ADR.		2026-06-04 02:47:03 -07:00
..
composition-capability-scope.md	docs: mark ADR-0174 Phase 5a shipped in composition-capability-scope	2026-06-03 16:23:53 -07:00
composition-wall-execution-plan-2026-06-03.md	feat(r4): flip cv-0005 to serving — train_sample 6/44/0 -> 7/43/0 (ADR-0207 §5 step 2)	2026-06-03 22:20:12 -07:00
comprehension-primitive-inventory.md	Add canonical composition analysis docs	2026-06-03 15:35:58 -07:00
extraction-richness-audit-2026-06-03.md	docs: ratify ADR-0207 — GSM8K comprehension/composition substrate	2026-06-03 19:42:47 -07:00
gsm8k-lift-program-strategy-2026-06-04.md	fix(gsm8k): disable unsound serving bridges — restore sealed wrong=0 (0/5 -> 0/0)	2026-06-04 01:55:05 -07:00
question-layer-gap-survey.md
real-gsm8k-capability-measurement-2026-06-04.md	docs: two lift attempts falsified + the ADR-0207 architectural impasse	2026-06-04 02:47:03 -07:00
solver-operation-coverage.md
wiring-promotion-gate-brief-2026-06-03.md	docs: composition-wall execution scope (ADR-0207 §5 step 2)	2026-06-03 21:14:02 -07:00