GRPO
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
2026.02 | 0.99 | 0.994 | 0.9 | 0.94 | 0.93 | 0.928 | 0.866 | 0.935 | |
2026.02 | 0.972 | 0.982 | 0.806 | 0.926 | 0.924 | 0.94 | 0.902 | 0.921 | |
2026.02 | 0.852 | 0.828 | 0.47 | 0.56 | 0.604 | 0.858 | 0.766 | 0.702 | |
2026.02 | 0.84 | 0.59 | 0.3 | 0.43 | 0.72 | 0.45 | 0.55 | 0.554 | |
2026.02 | 0.76 | 0.71 | 0.32 | 0.59 | 0.47 | 0.24 | 0.05 | 0.448 | |
2026.02 | 0.74 | 0.54 | 0.22 | 0.46 | 0.45 | 0.78 | 0.8 | 0.57 | |
2026.02 | 0.716 | 0.674 | 0.232 | 0.412 | 0.472 | 0.812 | 0.7 | 0.574 | |
2026.02 | 0.656 | 0.514 | 0.198 | 0.282 | 0.51 | 0.806 | 0.708 | 0.524 | |
2026.02 | 0.634 | 0.55 | 0.156 | 0.294 | 0.434 | 0.788 | 0.714 | 0.51 | |
2026.02 | 0.594 | 0.364 | 0.21 | 0.442 | 0.212 | 0.014 | 0.066 | 0.272 | |
2026.02 | 0.572 | 0.372 | 0.288 | 0.43 | 0.2 | 0.01 | 0.01 | 0.269 | |
2026.02 | 0.004 | 0 | 0.18 | 0.23 | 0 | 0 | 0 | 0.059 |