Static Multi-Session QA on LONGMEMEVAL S (noisy)
94.64SS-AOBLIVION
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| OBLIVIONModel=GPT-4o-mini, Method Type=ours2026.03 | 94.64 | 95.71 | 60 | 79.7 | 74.36 | 80.45 | 81.8 | |
| OBLIVIONModel=GPT-4.1-mini, Method Type=ours2026.03 | 94.64 | 95.71 | 96.67 | 84.96 | 93.59 | 88.72 | 90.6 | |
| LME-RFTModel=GPT-4o-mini, Method Type=direct2026.03 | 91.07 | 94.29 | 60 | 69.92 | 79.49 | 77.44 | 78.6 | |
| LME-RFTModel=GPT-4.1-mini, Method Type=direct2026.03 | 91.07 | 97.14 | 93.33 | 81.2 | 93.59 | 87.97 | 89 | |
| EVERMEMOSModel=GPT-4.1-mini, Method Type=memory2026.03 | 80.36 | 100 | 96.67 | 76.94 | 87.18 | 69.17 | 81.27 | |
| EVERMEMOSModel=GPT-4o-mini, Method Type=memory2026.03 | 55.36 | 91.43 | 93.33 | 60.4 | 67.95 | 53.38 | 65.47 |