| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| HaluMem | MemReader-0.6B | Memory Accuracy95.66 | 19 | 1mo ago | |
| HaluMem Medium | MemOS | R Score74.07 | 7 | 2mo ago | |
| LOCOMO (main) | MemReader-4B | Single Hop Score85.37 | 7 | 3mo ago | |
| BEHEMOTH in-distribution (test) | CluE | Personalization (MA)65.72 | 6 | 3mo ago |