Mathematical Reasoning on AMC 23 (Avg@8, Pass@8, Avg Response Length)
42.19Avg@8Full-vocabulary KL OPD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Full-vocabulary KL OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base2026.06 | 42.19 | 75 | 4,096 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=52026.06 | 40.31 | 70 | 2,010 | |
| Vanilla OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, Stabilization=Clip2026.06 | 38.75 | 72.5 | 2,097 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=102026.06 | 38.75 | 62.5 | 1,997 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=1002026.06 | 38.75 | 77.5 | 1,821 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=12026.06 | 37.5 | 70 | 2,113 | |
| Vanilla OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, Stabilization=Z-Score2026.06 | 36.56 | 67.5 | 2,087 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=0.52026.06 | 36.56 | 67.5 | 2,087 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=5002026.06 | 36.56 | 77.5 | 1,715 | |
| PowerOPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, alpha=0.12026.06 | 36.25 | 65 | 2,142 | |
| Vanilla OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base2026.06 | 35.94 | 62.5 | 1,881 | |
| Vanilla OPDTeacher=Qwen3-8B, Student=Qwen3-1.7B-Base, Stabilization=Tanh2026.06 | 35.31 | 62.5 | 1,996 |