Memory on RULER HotpotQA
67ScoreDARE + TIES
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| DARE + TIESModel Category=Merged Models2026.06 | 67 | — | — | |
| TIESModel Category=Merged Models2026.06 | 65 | — | — | |
| ISO-CModel Category=Merged Models2026.06 | 65 | — | — | |
| RESMERGEModel Category=Merged Models2026.06 | 65 | — | — | |
| TSV-MergeModel Category=Merged Models2026.06 | 65 | — | — | |
| AMPOModel Category=Base and Expert Models2026.06 | 64 | — | — | |
| TAModel Category=Merged Models2026.06 | 64 | — | — | |
| TSV-MergeModel Category=Merged Models2026.06 | 64 | — | — | |
| ISO-CTSModel Category=Merged Models2026.06 | 64 | — | — | |
| RAMModel Category=Merged Models2026.06 | 64 | — | — | |
| RAMModel Category=Merged Models2026.06 | 64 | — | — | |
| MemoryAgentModel Category=Base and Expert Models2026.06 | 63 | — | — | |
| TIESModel Category=Merged Models2026.06 | 63 | — | — | |
| TAModel Category=Merged Models2026.06 | 62 | — | — | |
| ISO-CModel Category=Merged Models2026.06 | 62 | — | — | |
| ToolRLModel Category=Base and Expert Models2026.06 | 61 | — | — | |
| AMPOModel Category=Base and Expert Models2026.06 | 61 | — | — | |
| ToolRLModel Category=Base and Expert Models2026.06 | 61 | — | — | |
| ISO-CTSModel Category=Merged Models2026.06 | 61 | — | — | |
| Qwen2.5-7B-InstructModel Category=Base and Expert Models2026.06 | 60 | — | — | |
| MemoryAgentModel Category=Base and Expert Models2026.06 | 60 | — | — | |
| DARE + TIESModel Category=Merged Models2026.06 | 60 | — | — | |
| RESMERGEModel Category=Merged Models2026.06 | 60 | — | — | |
| Qwen2.5-7B-InstructModel Category=Base and Expert Models2026.06 | 59 | — | — | |
| BaseModel Category=Base and Task Models2026.01 | — | 60.94 | 50 | |
| CURE (Coding)Model Category=Base and Task Models2026.01 | — | 58.59 | 56.25 | |
| DARE+TAModel Category=Merged Models2026.01 | — | 76.56 | 76.56 | |
| DARE+TA2026.01 | — | 76.56 | 76.56 | |
| DARE+TIESModel Category=Merged Models2026.01 | — | 75 | 77.34 | |
| FisherModel Category=Merged Models2026.01 | — | 60.06 | 49.22 | |
| MemAgent (Memory)Model Category=Base and Task Models2026.01 | — | 78.91 | 78.12 | |
| RAMModel Category=Merged Models2026.01 | — | 75.78 | 75 | |
| RAM2026.01 | — | 75.78 | 75 | |
| RAM+Model Category=Merged Models2026.01 | — | 79.69 | 78.13 | |
| RAM+2026.01 | — | 79.69 | 78.13 | |
| TAModel Category=Merged Models2026.01 | — | 69.53 | 68.75 | |
| TIESModel Category=Merged Models2026.01 | — | 71.88 | 82.03 | |
| ToolRL (Tool)Model Category=Base and Task Models2026.01 | — | 58.59 | 48.44 | |
| WUDI2026.01 | — | 78.12 | 78.91 |