Conversation
Remove excessive println! debug statements throughout the LlmPilot implementation to reduce noise and improve performance. Replace them with appropriate debug!/info! logging macros for better control. perf(evaluate): add per-node content caching to avoid recomputation Introduce node_content_cache in PipelineContext to store computed content for each node during aggregate_content phase. This allows build_response to reuse cached content instead of performing duplicate leaf traversal operations, improving performance. feat(search): implement L2/L3 cache boosting for search results Add L2 cache boost that uses historical path success data to adjust cue confidences based on document structure patterns. Implement L3 cache boost that refines candidate scores using cached strategy scores and stores updated scores for future queries. refactor(strategy): remove unused semantic strategy module Remove the SemanticStrategy implementation as it was no longer being used in the codebase. Update documentation and related references to reflect the removal of the semantic strategy option. feat(llm-strategy): add batch evaluation support for multiple nodes Implement batch evaluation capability in LlmStrategy where multiple nodes are scored in a single LLM call instead of individual calls. This reduces latency from O(N) to O(1) for node evaluation. Add appropriate batch prompt building and response parsing logic with fallback to single evaluation when needed.
- Include document title in keyword scoring alongside content - Implement shared pipeline budget controller for unified token consumption tracking across pilots - Add depth-first traversal in cross-document strategy for better node exploration - Introduce subtree searching with configurable depth limits - Add deduplication of scored nodes in cross-document strategy
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
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.