SaFeR-ToolKit (+ SFT+DPO+GRPO) [3B]
| Method | Links | |||
|---|---|---|---|---|
| 91.89 | 85.09 | 87.36 | ||
| 91.89 | 89.25 | 90.94 | ||
2026.03 | 76.42 | 73.21 | 40 | |
2026.03 | 46.6 | 54.53 | 11.89 | |
2026.03 | 2.81 | 2.8 | 2.84 | |
2026.03 | 2.8 | 2.68 | 2.72 | |
2026.03 | 2.8 | 2.7 | 2.28 | |
2026.03 | 2.78 | 2.7 | 2.82 | |
2026.03 | 2.77 | 2.87 | 2.88 | |
2026.03 | 2.75 | 2.84 | 2.88 | |
2026.03 | 2.73 | 2.4 | 2.46 | |
2026.03 | 2.72 | 2.65 | 2.2 | |
2026.03 | 2.72 | 2.72 | 2.86 | |
2026.03 | 2.68 | 2.65 | 2.78 | |
2026.03 | 2.62 | 2.35 | 2.5 | |
2026.03 | 2.62 | 2.31 | 2.27 | |
2026.03 | 2.6 | 2.54 | 2.67 | |
2026.03 | 2.52 | 2.46 | 2.63 | |
2026.03 | 2.49 | 1.67 | 1.8 | |
2026.03 | 2.46 | 2.01 | 1.94 | |
2026.03 | 2.43 | 1.63 | 1.74 | |
2026.03 | 2.37 | 2.64 | 2.27 | |
2026.03 | 2.27 | 2.66 | 2.27 | |
2026.03 | 1.7 | 1.63 | 1.38 | |
2026.03 | 1.39 | 2.16 | 1.66 | |
2026.03 | 1.38 | 2.35 | 1.73 |