Mathematical Reasoning on MATH500 (Avg@8, Pass@8, Length)
46.53Avg@8Full-vocab
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Full-vocabTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Full-vocab2026.06 | 46.53 | 54.38 | 4,095 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=102026.06 | 45.49 | 67 | 1,308 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=52026.06 | 45.33 | 66.72 | 1,348 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=0.52026.06 | 45.07 | 66.98 | 1,371 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=1002026.06 | 45.03 | 65.42 | 1,322 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=502026.06 | 44.91 | 65.78 | 1,313 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=12026.06 | 44.89 | 67.1 | 1,408 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=5002026.06 | 44.59 | 65.42 | 1,321 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=0.12026.06 | 43.82 | 65.86 | 1,395 | |
| Vanilla + ClipTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=+ Clip2026.06 | 43.75 | 65.3 | 1,429 | |
| Vanilla + TanhTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=+ Tanh2026.06 | 43.39 | 64.86 | 1,425 | |
| Vanilla + Z-ScoreTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=+ Z-Score2026.06 | 42.72 | 65.62 | 1,364 | |
| VanillaTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla2026.06 | 40.43 | 64 | 1,450 |