Text-based Question Answering on MuSiQue (EM, F1, R@k, #Tok)
3.6EM
Latent Memory
Updated 1mo ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.06 | 3.6 | 14.7 | 28.8 | 76 | |
2026.06 | 2.8 | 14.9 | - | 3,012 | |
2026.06 | 2.3 | 11.8 | 8.4 | 36 | |
2026.06 | 2.2 | 12.8 | 14.9 | 46 | |
2026.06 | 2 | 14.3 | 30.8 | 280 | |
2026.06 | 1.7 | 9.5 | 12.1 | 119 | |
2026.06 | 1.7 | 10.7 | 22.4 | 250 | |
2026.06 | 1.6 | 12.8 | 31.3 | 247 | |
2026.06 | 1.6 | 13.4 | 34.8 | 262 | |
2026.06 | 1.5 | 11.2 | 18.5 | 126 | |
2026.06 | 1.4 | 9.5 | - | 42 | |
2026.06 | 1.4 | 12.1 | 16.5 | 133 | |
2026.06 | 1.3 | 10.3 | - | 708 | |
2026.06 | 1.3 | 11.4 | 9.8 | 83 | |
2026.06 | 1.2 | 11.1 | 16.9 | 119 | |
2026.06 | 1 | 10.2 | 9.5 | 77 | |
2026.06 | 1 | 10.6 | 10.3 | 81 | |
2026.06 | 0.9 | 7.4 | - | 371 | |
2026.06 | 0.9 | 8.1 | 6.9 | 75 | |
2026.06 | 0.7 | 6 | 22.3 | 119 | |
2026.06 | 0.5 | 5.5 | - | 191 |