Baseline + pre-warming + ES Text QA (w/ CoT)
| Method | Links | ||
|---|---|---|---|
| 60.4 | 96.25 | ||
2026.02 | 55 | - | |
| 52.39 | 91.85 | ||
2026.02 | 51.6 | - | |
2026.02 | 51.5 | - | |
2026.02 | 51.11 | - | |
2026.02 | 50.02 | - | |
2026.02 | 50.01 | 82.58 | |
2026.02 | 49.8 | - | |
2026.02 | 47.53 | - | |
2026.02 | 47.4 | 85.75 | |
2026.02 | 45.4 | - | |
2026.02 | 45.2 | - | |
2026.02 | 43.6 | - | |
2026.02 | 40.5 | - | |
2026.02 | 40.2 | - | |
2026.02 | 39.1 | - | |
2026.02 | 22.03 | - |