GitNexus/eval/workflow_bench/oracles
Gergő Magyar 2251adeeb2
fix(eval): grade real C include provenance (#3611)
* fix(eval): verify independent C repairs and reject broken controls

* fix(eval): replace invalid C task with provenance behavior

* fix(eval): grade real C include provenance

* test(eval): add independent typed C reference repairs

* docs(eval): record C oracle evidence contract

* fix(eval): remove invalid synthetic C oracle

* fix(eval): remove type-invalid C reference

* docs: remove C oracle planning document

* test(eval): tolerate formatting in C oracle mutants

* test(eval): clarify C oracle node and file bindings
2026-10-17 15:03:42 +01:00
..
reference fix(eval): grade real C include provenance (#3611) 2026-10-17 15:03:42 +01:00
cross-module-parse-retry.oracle.test.ts fix(release): gate publication on accuracy and paired evaluation (#3503) 2026-10-10 08:26:08 +01:00
inv-feature-c-include-provenance.oracle.test.ts fix(eval): grade real C include provenance (#3611) 2026-10-17 15:03:42 +01:00
inv-feature-list-repos-filter.oracle.test.ts fix(release): gate publication on accuracy and paired evaluation (#3503) 2026-10-10 08:26:08 +01:00
review-pr-2108-defect.labels.json fix(eval): make evolution evidence valid and bounded 2026-09-05 10:08:35 +00:00
review-pr-2258-clean.labels.json fix(eval): make evolution evidence valid and bounded 2026-09-05 10:08:35 +00:00
review-pr-2258-defect.labels.json fix(eval): make evolution evidence valid and bounded 2026-09-05 10:08:35 +00:00
review-pr-2718-defect.labels.json fix(eval): make evolution evidence valid and bounded 2026-09-05 10:08:35 +00:00
review-pr-2773-clean.labels.json fix(eval): make evolution evidence valid and bounded 2026-09-05 10:08:35 +00:00
review-pr-2794-defect.labels.json fix(eval): make evolution evidence valid and bounded 2026-09-05 10:08:35 +00:00
trivial-status-json-alias.oracle.test.ts fix(release): gate publication on accuracy and paired evaluation (#3503) 2026-10-10 08:26:08 +01:00
vitest.config.mts fix(release): gate publication on accuracy and paired evaluation (#3503) 2026-10-10 08:26:08 +01:00