Long-term Memory Question Answering on ATM-Bench 1.0 (test)
77.8QS
Oracle
Updated 2mo ago
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
2026.05 | 77.8 | - | - | 85 | 90.4 | 69.5 | |
2026.05 | 70 | - | - | 81.8 | 69.3 | 61.9 | |
2026.05 | 52.5 | 70.3 | 51 | 57.2 | 50.4 | 48.8 | |
2026.05 | 51 | 68.7 | 48.6 | 60.3 | 32.4 | 48.2 | |
2026.05 | 50.3 | 68.7 | 48.2 | 59.7 | 35.3 | 48.4 | |
2026.05 | 48.4 | 66.1 | 46.9 | 55.3 | 45.3 | 42.8 | |
2026.05 | 47.7 | 69.6 | 46.9 | 59.4 | 39.3 | 41.8 | |
2026.05 | 46.1 | 66.6 | 44 | 55 | 27 | 44.9 | |
2026.05 | 44.8 | 66.4 | 42.8 | 57.2 | 16.9 | 43.6 | |
2026.05 | 43.5 | 61.9 | 41.8 | 57.2 | 25.9 | 48.4 | |
2026.05 | 42.9 | 66.4 | 41.5 | 58.6 | 34.9 | 34.1 | |
2026.05 | 42.1 | 61.8 | 42.8 | 46.4 | 33.1 | 30.9 | |
2026.05 | 42 | 61.8 | 41.3 | 50.6 | 31 | 38.9 | |
2026.05 | 41.5 | 61.9 | 38.4 | 58.6 | 25.9 | 33.7 | |
2026.05 | 0.2 | - | - | 0 | 0 | 0.6 |