D-SFT+DAPO+Short-RL+GroupAdapt (τ=0.5)
| Method | Links | ||||
|---|---|---|---|---|---|
| 97.3 | 97.3 | 99.39 | 0.2 | ||
2026.06 | 96.9 | 96.9 | 99.23 | 0 | |
2026.06 | 96.2 | 96.2 | 98.64 | 0.2 | |
2026.06 | 93.8 | 96.9 | 98.36 | 3.6 | |
2026.06 | 92.8 | 93.8 | 95.34 | 1 | |
2026.06 | 92.8 | 95.4 | 96.49 | 4.6 | |
2026.06 | 92.3 | 95.9 | 96.88 | 4.6 | |
2026.06 | 92.3 | 95.9 | 98.32 | 4.1 | |
2026.06 | 91.2 | 94.8 | 97.21 | 4.1 | |
2026.06 | 90.7 | 94.3 | 96.88 | 6.2 | |
2026.06 | 89.7 | 94.8 | 96.92 | 6.2 | |
2026.06 | 88.1 | 93.3 | 95.17 | 9.3 | |
2026.06 | 87.1 | 87.6 | 89.39 | 0.5 | |
2026.06 | 86.6 | 90.7 | 93.8 | 7.2 | |
2026.06 | 83 | 89.7 | 93.71 | 12.4 | |
2026.06 | 77.3 | 83 | 89.84 | 13.4 | |
2026.06 | 76.3 | 90.7 | 93.14 | 19.1 |