Mathematical Reasoning on Minerva (Avg@8, Pass@8, average response length)
11.76Avg@8PowerOPD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=102026.06 | 11.76 | 24.63 | 1,056 | |
| Full-vocabTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Full-vocab2026.06 | 11.58 | 15.07 | 4,089 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=5002026.06 | 11.44 | 23.9 | 1,309 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=12026.06 | 11.31 | 25.37 | 1,188 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=1002026.06 | 11.31 | 22.79 | 1,244 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=502026.06 | 11.12 | 23.9 | 1,149 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=52026.06 | 11.08 | 22.79 | 1,076 | |
| Vanilla + Z-ScoreTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=+ Z-Score2026.06 | 10.71 | 21.32 | 1,212 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=0.12026.06 | 10.25 | 22.43 | 1,195 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=0.52026.06 | 10.06 | 22.79 | 1,144 | |
| Vanilla + TanhTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=+ Tanh2026.06 | 9.65 | 21.32 | 1,281 | |
| Vanilla + ClipTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=+ Clip2026.06 | 9.42 | 22.06 | 1,286 | |
| VanillaTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla2026.06 | 9.01 | 20.59 | 1,306 |