Feature/lexicon multiplier retune#102
Merged
Merged
Conversation
Config.lexicon_boost_clamp Replace the hardcoded clamp(-1.0, 2.0) — a query-independent importance multiplier of up to 3x — with Config.lexicon_boost_clamp (default 0.05). Fused RRF relevance scores are tightly compressed (rank-1-to-2 gap ~1-2%), so a wide bound let the lexicon's query-independent importance signal overwhelm relevance and bury the evidence. On LongMemEval the old bound dropped single-session Recall@1 from 0.94 to 0.38 even with semantic search active; tuning the clamp showed a smooth tradeoff converging to the no-lexicon baseline only as c -> 0. It is a relevance<->importance dial, not a bug to tune away, so it is now exposed with a conservative default: 0.05 recovers ~87% of the no-lexicon Recall@1 while still expressing modest importance. 0.0 disables lexicon influence; larger values weight importance more heavily. Value validated non-negative/finite in ContextEngine::new.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
No description provided.