feat(recall): semantic stemming — paginate↔pagination, authenticate↔authentication#85
Merged
Conversation
…ion, authenticate↔authentication recall_approach now matches lexically-different but semantically-equal terms by collapsing words to a common root before the TF-cosine ranking: a query about "paginate" recalls a past episode about "pagination"; "authenticate" recalls "authentication"; "backups" recalls "backup". - approach-recall.ts: stem() (PURE, longest-suffix-first, guards words ≤4 chars) + semanticTokens(); the ranker vectors both query and sources through it. Localized to recall — the shared tokenizer (skill-dedup thresholds) is untouched. - Composes with the existing recency tilt + MMR diversity.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Strengthens
recall_approach's search from exact-token to semantic-lexical: it now matches surface-different but same-root terms by stemming before the TF-cosine ranking.Implementation:
approach-recall.ts:stem()(PURE, longest-suffix-first, guards words ≤4 chars) +semanticTokens(); the ranker vectors both query and sources through ittokenize(which drives skill-dedup thresholds) is deliberately untouched, so no existing similarity behavior shifts+2 tests (semantic match end-to-end;
stem()unit cases). Full suite 1448 green, tsc clean.