Mathematical Reasoning on AMC 23 (Avg@8, Pass@8, Response Length)
29.06Avg@8PowerOPD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=1002026.06 | 29.06 | 57.5 | 2,038 | |
| Vanilla + ClipTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=+ Clip2026.06 | 28.12 | 52.5 | 2,538 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=502026.06 | 28.12 | 57.5 | 2,088 | |
| Full-vocabTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Full-vocab2026.06 | 27.5 | 35 | 4,096 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=0.52026.06 | 25.94 | 55 | 2,535 | |
| Vanilla + Z-ScoreTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=+ Z-Score2026.06 | 25.62 | 50 | 2,269 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=52026.06 | 25.62 | 55 | 2,503 | |
| Vanilla + TanhTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=+ Tanh2026.06 | 25 | 57.5 | 2,468 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=102026.06 | 24.38 | 60 | 2,251 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=0.12026.06 | 24.06 | 55 | 2,460 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=12026.06 | 23.44 | 52.5 | 2,560 | |
| VanillaTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=Vanilla, Param=–2026.06 | 22.19 | 50 | 2,393 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-0.6B-Base, Class=PowerOPD, alpha=5002026.06 | 22.19 | 57.5 | 1,952 |