Robust-R1 (SFT and RL)
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
2025.12 | 65.29 | 63.91 | 60.97 | 49.14 | 49.09 | 49.8 | 40.68 | 37.81 | 34.84 | 50.17 | |
2025.12 | 64.11 | 60.22 | 57.32 | 48.72 | 48.54 | 49.04 | 37.78 | 37.04 | 33.3 | 48.45 | |
2025.12 | 62.35 | 60.24 | 59.14 | 49.82 | 45.39 | 51.08 | 36.67 | 30.41 | 28.51 | 47.06 | |
2025.12 | 61.76 | 60.87 | 56.1 | 48.04 | 48.36 | 50.12 | 40.8 | 38.58 | 35.18 | 48.86 | |
2025.12 | 58.23 | 57.76 | 50.6 | 48.65 | 46.3 | 44.19 | 40.48 | 37.46 | 34.8 | 46.49 | |
2025.12 | 47.05 | 46.58 | 40.24 | 45.03 | 43.39 | 47.43 | 22.9 | 22.19 | 19.83 | 37.18 | |
2025.12 | 46.47 | 42.23 | 40.24 | 46.87 | 39.94 | 44.61 | 21.11 | 21.95 | 19.37 | 35.86 | |
2025.12 | 33.52 | 26.08 | 30.48 | 26.07 | 22.12 | 24.43 | 0.68 | 0.65 | 0.67 | 18.3 |