core/generate/derivation
Shay 92a7106ad0 feat(derivation): Workstream A inc 1 — reader EX passes (half-of/fraction/comparatives) + age hygiene + 30+ Phase B exemplars + proxy rebaseline (ratified)
Governing ratifications (written first):
- docs/analysis/problem-solving-lift-strategic-deep-dive-ratification-2026-06-16.md
- docs/analysis/gsm8k-derivation-reader-recognizer-lift-workstream-a-ratification-2026-06-17.md

- generate/derivation/extract.py: lexeme EX passes for fractions/comparatives + refined incidental-only age hygiene (preserves pinned tests + helps proxy cases).
- teaching/admissibility_exemplars/*_v1.jsonl: 30+ new phase_b_seed entries (dcs/rwc/ma/ca/ta/dsnq) with full schema + Workstream A provenance.
- evals/gsm8k_math/train_sample/v1/report.json: rebaselined via runner; 6/44/0 (wrong=0).
- docs/analysis/*ratification*.md + lookback-2026-06-17.md included.

Scope per ratif: reader + exemplar growth + rebaseline + lookback. No sealed, no CLOSE (deferred per posture), 0 wrong preserved.

See ratifs + lookback for full verification obligations, evidence, and test plan.
2026-06-16 20:56:58 -07:00
..
state test(derivation): add ADR-0184 semantic replay equivalence harness 2026-06-10 14:35:22 -07:00
__init__.py feat(derivation): define semantic-ledger candidate source boundary 2026-06-10 13:17:53 -07:00
accumulate.py feat(derivation): define semantic-ledger candidate source boundary 2026-06-10 13:17:53 -07:00
clauses.py feat(adr-0178-gb1): clause segmentation + clause-local sub-derivation 2026-05-28 17:19:50 -07:00
comparatives.py feat(adr-0176-ms2): multi-step chain model — text + comparative operands 2026-05-28 16:35:41 -07:00
compose.py feat(adr-0178-gb3a): clause-scoped referent guard — refuse cross-clause aggregation (#456) 2026-05-29 09:15:52 -07:00
extract.py feat(derivation): Workstream A inc 1 — reader EX passes (half-of/fraction/comparatives) + age hygiene + 30+ Phase B exemplars + proxy rebaseline (ratified) 2026-06-16 20:56:58 -07:00
goal_residual.py 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
model.py feat(adr-0176-ms2): multi-step chain model — text + comparative operands 2026-05-28 16:35:41 -07:00
multistep.py feat(adr-0177-cp2a): cue-precision ledger training + measurement (+ unit hygiene) (#461) 2026-05-29 10:21:58 -07:00
pool.py feat(derivation): define semantic-ledger candidate source boundary 2026-06-10 13:17:53 -07:00
product_bridge.py feat(adr-0195): GSM8K product promotion bridge — serving 4/46/0 → 6/44/0, wrong=0 (#500) 2026-05-30 17:33:56 -07:00
r1_reconstruction.py feat(derivation): define semantic-ledger candidate source boundary 2026-06-10 13:17:53 -07:00
search.py feat(adr-0182): cross-composer disagreement pooling — distractor 0014 + disguised-polarity refuse (confuser wrong 5->2) (#476) 2026-05-29 13:22:19 -07:00
target.py feat(adr-0182): prior-state question guard — temporal-scope confuser refuses (confuser wrong 2->1, pair-tells ->0) (#480) 2026-05-29 14:22:28 -07:00
verify.py feat(derivation): math-serving reach seam — select_self_verified is policy-aware (ADR-0206 §5) 2026-06-06 15:46:14 -07:00