Preference Evaluation via Multi-Choice Queries on PrefEval Explicit
81.3AccuracyMemCoE
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MemCoEBackbone=Qwen2.5-7B-Instruct, Retrieval model=all-MiniLM-L6-v2, Top-K retrieval=102026.05 | 81.3 | 52.02 | |
| MemAgentBackbone=Qwen2.5-7B-Instruct, Retrieval model=all-MiniLM-L6-v2, Top-K retrieval=102026.05 | 72.3 | 45 | |
| Mem-αBackbone=Qwen3-4B, Retrieval model=all-MiniLM-L6-v2, Top-K retrieval=102026.05 | 71.9 | 44.19 | |
| LightMemBackbone=Qwen2.5-7B-Instruct, Retrieval model=all-MiniLM-L6-v2, Top-K retrieval=102026.05 | 64.2 | 41.21 | |
| A-MemBackbone=Qwen2.5-7B-Instruct, Retrieval model=all-MiniLM-L6-v2, Top-K retrieval=102026.05 | 62.3 | 42.64 | |
| Mem0Backbone=Qwen2.5-7B-Instruct, Retrieval model=all-MiniLM-L6-v2, Top-K retrieval=102026.05 | 57.6 | 38.23 | |
| RAGBackbone=Qwen2.5-7B-Instruct, Retrieval model=all-MiniLM-L6-v2, Top-K retrieval=102026.05 | 47.8 | 36.68 | |
| Long ContextBackbone=Qwen2.5-7B-Instruct2026.05 | 31.7 | 26.9 |