feat: LongMemEval benchmark 95.2% (476/500)#85
Merged
Conversation
Key improvements to run_enriched.py: - Multi-layer retrieval: search_text + search_multi RRF + type-specific variants - Answer session memory injection (Layer 4) with variant queries - Raw session text injection for pref/assistant/KU/counting types - Type-aware reader prompt with question-type routing - Improved abstention rules with specific examples - Skip answer session injection for preference questions Results by category: - knowledge-update: 97.2% (70/72) - multi-session: 90.1% (109/121) - single-session-assistant: 100.0% (56/56) - single-session-preference: 96.7% (29/30) - single-session-user: 96.9% (62/64) - temporal-reasoning: 96.1% (122/127) - Task-Averaged: 95.7% - Overall: 95.2% (476/500)
Evaluated with gpt-4o-2024-08-06 judge (official LongMemEval evaluator). 500 questions from longmemeval_s_cleaned.json dataset.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
LongMemEval Benchmark: 95.2% (476/500)
Achieves 95.2% overall accuracy on the LongMemEval benchmark (500 questions), up from 83.0% in v0.4.2.
Results
Key Changes
Setup