Add Wiki knowledge evaluation - #14
Merged
Merged
Conversation
Gate retrieval with reviewed real-question cases and source metrics. Fix mixed Korean suffix handling and multi-term query noise. Validation: pnpm check
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Purpose
Add a deterministic, reviewed evaluation gate for LLM Wiki knowledge
retrieval so search behavior can be changed without silently regressing
real questions or unrelated-query abstention.
Changes
expectations
abstention accuracy
pnpm checkisolated matches in multi-term queries
Impact
Knowledge retrieval remains local, deterministic, provider-neutral, and
API-free. Embeddings and graph retrieval stay deferred until observed
misses justify the additional complexity.
Validation
pnpm check