Mathematical Reasoning on AIME 25 (Avg@8, Pass@8, Avg Length)
7.08Avg@8Vanilla OPD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Vanilla OPDTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base2026.06 | 7.08 | 13.33 | 3,282 | |
| PowerOPDTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base, alpha=0.12026.06 | 7.08 | 23.33 | 3,081 | |
| Vanilla OPD + TanhTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base2026.06 | 6.67 | 16.67 | 3,240 | |
| PowerOPDTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base, alpha=5002026.06 | 6.67 | 20 | 2,588 | |
| PowerOPDTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base, alpha=1002026.06 | 6.25 | 23.33 | 2,732 | |
| Full-vocab KL OPDTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base2026.06 | 5.83 | 20 | 4,096 | |
| PowerOPDTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base, alpha=102026.06 | 5 | 13.33 | 3,188 | |
| Vanilla OPD + ClipTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base2026.06 | 4.58 | 20 | 3,381 | |
| Vanilla OPD + Z-ScoreTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base2026.06 | 4.17 | 13.33 | 3,244 | |
| PowerOPDTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base, alpha=12026.06 | 4.17 | 16.67 | 3,136 | |
| PowerOPDTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base, alpha=52026.06 | 4.17 | 13.33 | 3,283 | |
| PowerOPDTeacher=Qwen3-4B, Student=Qwen3-1.7B-Base, alpha=502026.06 | 4.17 | 13.33 | 2,876 |