lukaszraczylo 9d1d4d89e9 fix(catalog): unify negation guard and harden round-2 evaluators ...

A shared NoUnnegatedMention primitive replaces four per-category
implementations, with a bidirectional clause-scoped window and an
adjectival cue set. The docker layer fixture drops WORKDIR, which
creates a real layer and made the pinned answer wrong. The replica
failover test now forces a JSON verdict so a negated answer cannot
score. Remaining exact-answer tests migrate to ExactToken. Ordering
prompts shuffle their candidate lists so echoing scores zero.

2026-08-17 03:50:57 +01:00

..

composite_test.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

composite.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

contains_test.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

contains.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

equals_test.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

equals.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

eval.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

exacttoken_test.go

fix(catalog): harden evaluators against correct-answer phrasing variance

2026-08-17 02:48:05 +01:00

exacttoken.go

fix(catalog): harden evaluators against correct-answer phrasing variance

2026-08-17 02:48:05 +01:00

exec_test.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

exec.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

extract_test.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

extract.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

jsonarray.go

fix: floor token budgets and surface truncation, add live progress

2026-08-17 01:55:14 +01:00

jsonfield_test.go

fix(catalog): unify negation guard and harden round-2 evaluators

2026-08-17 03:50:57 +01:00

jsonfield.go

fix(catalog): unify negation guard and harden round-2 evaluators

2026-08-17 03:50:57 +01:00

negation_test.go

fix(catalog): unify negation guard and harden round-2 evaluators

2026-08-17 03:50:57 +01:00

negation.go

fix(catalog): unify negation guard and harden round-2 evaluators

2026-08-17 03:50:57 +01:00

numeric_test.go

fix(catalog): unify negation guard and harden round-2 evaluators

2026-08-17 03:50:57 +01:00

numeric.go

fix(catalog): unify negation guard and harden round-2 evaluators

2026-08-17 03:50:57 +01:00

regex_test.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00

regex.go

feat: LLM accuracy testing framework with 17-test catalog

2026-08-17 01:29:45 +01:00