Mathematical Reasoning on AIME 25 (P@1, P@4, P@8)
83.33P@8MOTAB
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MOTABTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct2026.05 | 83.33 | 74.14 | 47.5 | |
| SKDTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct, beta=0.82026.05 | 80 | 72.52 | 45 | |
| SKDTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct, beta=1.02026.05 | 80 | 64.33 | 33.75 | |
| SFTTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct2026.05 | 73.33 | 61.48 | 37.5 | |
| SKDTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct, beta=0.62026.05 | 66.67 | 61.48 | 39.58 | |
| ImitKDTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct2026.05 | 63.33 | 57.29 | 37.92 | |
| MOTABTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct2026.05 | 56.67 | 50.71 | 31.67 | |
| SKDTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct, beta=0.62026.05 | 46.67 | 43.33 | 25 | |
| SKDTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct, beta=0.82026.05 | 46.67 | 45 | 27.5 | |
| SKDTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct, beta=1.02026.05 | 46.67 | 40 | 25.83 | |
| SFTTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct2026.05 | 40 | 35.43 | 24.17 | |
| RFTTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct2026.05 | 40 | 30.95 | 21.67 | |
| ImitKDTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct2026.05 | 40 | 33.05 | 20 |