Question Answering on RULER HQA
66.86AccuracyMHM-LRU
Evaluation Results
| Method | Links | |
|---|---|---|
| MHM-LRUContext Length=7K2026.07 | 66.86 | |
| ReMemContext Length=7K2026.07 | 64.32 | |
| MHM-LRUContext Length=14K2026.07 | 62.76 | |
| MemAgentContext Length=7K2026.07 | 62.24 | |
| LLMContext Length=14K2026.07 | 60.94 | |
| MemAgentContext Length=14K2026.07 | 60.94 | |
| LLMContext Length=7K2026.07 | 60.16 | |
| ReMemContext Length=14K2026.07 | 59.64 | |
| MHM-LRUContext Length=28K2026.07 | 59.64 | |
| LLMContext Length=56K2026.07 | 57.03 | |
| MemAgentContext Length=28K2026.07 | 56.25 | |
| MHM-LRUContext Length=56K2026.07 | 55.21 | |
| MemAgentContext Length=56K2026.07 | 50.52 | |
| ReMemContext Length=28K2026.07 | 50.26 | |
| MHM-LRUContext Length=112K2026.07 | 50.26 | |
| MHM-LRUContext Length=224K2026.07 | 50.26 | |
| LLMContext Length=28K2026.07 | 50 | |
| LLMContext Length=112K2026.07 | 50 | |
| MHM-LRUContext Length=896K2026.07 | 49.74 | |
| MHM-LRUContext Length=448K2026.07 | 48.7 | |
| MemAgentContext Length=224K2026.07 | 42.97 | |
| MemAgentContext Length=112K2026.07 | 42.19 | |
| ReMemContext Length=56K2026.07 | 39.06 | |
| LLMContext Length=224K2026.07 | 37.5 | |
| MemAgentContext Length=448K2026.07 | 33.33 | |
| ReMemContext Length=112K2026.07 | 29.68 | |
| ReMemContext Length=224K2026.07 | 26.56 | |
| MemAgentContext Length=896K2026.07 | 21.62 | |
| ReMemContext Length=448K2026.07 | 16.14 | |
| ReMemContext Length=896K2026.07 | 13.8 | |
| LLMContext Length=448K2026.07 | 8.59 | |
| LLMContext Length=896K2026.07 | 0 |