Question Answering on RULER-HQA 224K context length
79.69Normalized Sub-EM AccuracyLycheeMemory-7B w/o Gate
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LycheeMemory-7B w/o GateModel=LycheeMemory-7B, Gate=disabled2026.02 | 79.69 | — | |
| LycheeMemory-7BModel=LycheeMemory-7B, Gate=enabled2026.02 | 73.44 | 5.9 | |
| RL-MemAgent-7BModel=RL-MemAgent-7B2026.02 | 72.66 | — | |
| RAG + Qwen2.5-7B-InstructModel=RAG + Qwen2.5-7B-Instruct2026.02 | 64.06 | — | |
| Search-R1Model=Search-R12026.02 | 62.5 | — | |
| Qwen2.5-Instruct-14B-1MModel=Qwen2.5-Instruct-14B-1M2026.02 | 37.5 | — | |
| Qwen2.5-Instruct-7B-1MModel=Qwen2.5-Instruct-7B-1M2026.02 | 33.59 | — | |
| QwenLong-L1-32BModel=QwenLong-L1-32B2026.02 | 17.19 | — | |
| DS-Distill-Qwen-32BModel=DS-Distill-Qwen-32B2026.02 | 13.28 | — | |
| DS-Distill-Qwen-14BModel=DS-Distill-Qwen-14B2026.02 | 8.59 | — | |
| DS-Distill-Qwen-7BModel=DS-Distill-Qwen-7B2026.02 | 0.78 | — |