Mathematical Reasoning on AIME 24 (P@8, P@4, P@1)
86.67P@8MOTAB
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MOTABTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct2026.05 | 86.67 | 78.33 | 47.08 | |
| SKDTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct, beta=0.82026.05 | 83.33 | 75.29 | 46.66 | |
| SKDTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct, beta=0.62026.05 | 80 | 74.57 | 45.83 | |
| SKDTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct, beta=1.02026.05 | 80 | 73.14 | 36.25 | |
| SFTTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct2026.05 | 76.67 | 73.14 | 43.33 | |
| ImitKDTeacher=gpt-oss-120b, Student=Qwen3-4B-Instruct2026.05 | 76.67 | 72.29 | 44.17 | |
| MOTABTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct2026.05 | 66.67 | 57.57 | 34.17 | |
| SKDTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct, beta=1.02026.05 | 56.67 | 43.33 | 27.5 | |
| SKDTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct, beta=0.62026.05 | 53.33 | 45 | 25 | |
| SKDTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct, beta=0.82026.05 | 53.33 | 48.33 | 24.17 | |
| SFTTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct2026.05 | 50 | 37.81 | 21.25 | |
| RFTTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct2026.05 | 46.67 | 38.38 | 21.67 | |
| ImitKDTeacher=Qwen3-32B, Student=Qwen3-4B-Instruct2026.05 | 46.67 | 40.71 | 24.17 |