Mathematical Problem Solving on AIME25 (mean@32 accuracy)
46.7Mean@32 AccuracyTeacher Model
Evaluation Results
| Method | Links | |
|---|---|---|
| Teacher ModelModel=Qwen3-4B-Instruct-2507, Teacher-Student Overlap=Large2026.06 | 46.7 | |
| IW-OPD10Student Model=Qwen3-4B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 46.7 | |
| IW-OPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 46.7 | |
| OPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 46.3 | |
| OPD10Student Model=Qwen3-4B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 45.7 | |
| IW-OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 27.1 | |
| OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 26.4 | |
| IW-OPD10Student Model=Qwen3-1.7B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 25.9 | |
| OPD10Student Model=Qwen3-1.7B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 24.7 | |
| BaseStudent Model=Qwen3-4B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 21.4 | |
| IW-OPD10Student Model=Qwen3-0.6B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 19 | |
| IW-OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 19 | |
| OPD10Student Model=Qwen3-0.6B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 17.1 | |
| OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 17.1 | |
| BaseStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 11 | |
| BaseStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 3.4 |