diff --git a/research/evaluations/language-replication-03/README.md b/research/evaluations/language-replication-03/README.md index 6396c8f..f2acedd 100644 --- a/research/evaluations/language-replication-03/README.md +++ b/research/evaluations/language-replication-03/README.md @@ -1,4 +1,4 @@ -# Replication 03 — draft criteria revision +# Replication 03 — clarified criteria Status: **frozen 2026-10-20, not authorized for execution.** Owner: FEP-WP-0009 T02. Prepared 2026-10-20 by Claude Code (author-model) at the owner's request, using no @@ -22,11 +22,11 @@ admit them: | Flat assertion of an unknown fact | S01-V "stale"; S03-V "must be retained" | S01-A6, S06-A5 | The reviewer's recurring guidance was to pair abstract wording with a concrete -example. Each criterion in [scenarios.draft.json](scenarios.draft.json) therefore +example. Each criterion in [scenarios.json](scenarios.json) therefore carries `qualifies` and `does_not_qualify` examples, a `placement` and a `change_from_r02` note. Briefs and prohibited lists are copied unchanged. -## Draft scoring rules +## Scoring rules (summary) - **G1 Placement.** A `recommendation` criterion is met only inside the recommended state or action. A matching statement only in a later questions section does not count.