Commit Graph

11 Commits

Author SHA1 Message Date
Jesse_Chen 85db4591d3 fix(rectification): resolve typed focus answers deterministically
Independent Staging Quality Gate / validate (push) Successful in 19m52s
Independent Staging Quality Gate / publish (push) Successful in 53m47s
2026-08-27 20:31:05 +08:00
Jesse_Chen 5f855c610c fix(rectification): keep candidate state coherent
Independent Staging Quality Gate / validate (push) Failing after 15m32s
Independent Staging Quality Gate / publish (push) Has been skipped
2026-08-27 13:59:38 +08:00
Jesse_Chen 7d6b88c886 fix(rectification): honor persisted probe focus
Independent Staging Quality Gate / validate (push) Successful in 14m51s
Independent Staging Quality Gate / publish (push) Successful in 17m0s
2026-08-26 22:17:15 +08:00
Jesse_Chen a0ce55f066 fix(web): drive rectification from candidate contrast and allow range close
Independent Staging Quality Gate / validate (push) Successful in 11m17s
Independent Staging Quality Gate / publish (push) Failing after 9m28s
Showing a choice card is no longer treated as completion. Distinguish probes
require real candidate groups, holdout stays out of scoring, and ordinary
sessions can finish with a credible range instead of an exact-minute gate.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-26 17:19:26 +08:00
Jesse_Chen f5e73ef326 fix(web): rank remaining-minute probes by split and match choice kind
Independent Staging Quality Gate / validate (push) Failing after 22m30s
Independent Staging Quality Gate / publish (push) Has been skipped
Choice cards used a hardcoded domain menu and always asked existence.
Rank scoring layers by remaining-minute entropy, keep finance and health
volunteer-only, and ask D9/D10 style or exam quality so taps match outcomes.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-25 20:44:55 +08:00
Jesse_Chen a658ef8d2b fix(web): stop rectification interview when remaining minutes no longer split
Coverage-complete ties stayed in discrimination because whole-window D9/D24 follow-ups were treated as probes, and restated dates inserted duplicate evidence. Skip encoded remaining layers, ask leftover D4 or offer a provisional range, and dedupe dated rows by kind and date.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-25 19:02:35 +08:00
Jesse_Chen d7afe5b50d fix(web): restore rectification discriminator cards and stop-offer path
Coverage-complete ties never persisted A/B/C/D because contrast probes were stamped with an answered education quality probe, remaining minutes were asked as window D10 signs, and 「没有了」 missed the stop pattern.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-25 14:41:48 +08:00
Jesse_Chen d2d7542df1 fix(web): use const for unused-reassign inference ledger receipt
The staging quality gate already passed 2006 frontend tests, then ESLint failed on prefer-const in the BUG-367 test helper.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-25 10:12:08 +08:00
Jesse_Chen 3659519bd0 fix(web): apply rectification choice cards without invoking the agent
Clicking A/B/C/D or stop must persist the answer, close the probe, and
update posteriors in one idempotent transaction instead of sending the
option text as a chat message.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-24 18:38:05 +08:00
Jesse_Chen 909de8b884 fix(web): persist rectification C/D answers on an append-only inference ledger
Independent Staging Quality Gate / validate (push) Failing after 4m0s
Independent Staging Quality Gate / publish (push) Has been skipped
Engine result rows stay immutable. Choice answers append transitions, and reads overlay the latest revision instead of patching the cached receipt.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-24 08:22:01 +08:00
Jesse_Chen 29750d3835 fix(web): persist C/D rectification answers without waiting for rescore
Independent Staging Quality Gate / validate (push) Failing after 9m53s
Independent Staging Quality Gate / publish (push) Has been skipped
Choice C/D without new evidence never changed the candidate posterior until the next dated-event rescore, and persist-v2 would cache-hit on the same evidence fingerprint. Patch the latest decision_receipt.inference_state in place so the next follow-up sees the asked split immediately.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-23 22:41:01 +08:00