Mathematical Problem Solving on AIME 24 (mean@32 accuracy)
60.4Mean@32 AccuracyTeacher Model
Evaluation Results
| Method | Links | |
|---|---|---|
| Teacher ModelModel=Qwen3-4B-Instruct-2507, Teacher-Student Overlap=Large2026.06 | 60.4 | |
| IW-OPD10Student Model=Qwen3-4B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 58.7 | |
| IW-OPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 58.7 | |
| OPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 56.5 | |
| OPD10Student Model=Qwen3-4B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 56.3 | |
| IW-OPD10Student Model=Qwen3-1.7B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 35.2 | |
| IW-OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 35.2 | |
| OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 34 | |
| OPD10Student Model=Qwen3-1.7B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 33.4 | |
| BaseStudent Model=Qwen3-4B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 23.1 | |
| IW-OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 13.6 | |
| BaseStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 13.4 | |
| IW-OPD10Student Model=Qwen3-0.6B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 12.4 | |
| OPDStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 11.8 | |
| OPD10Student Model=Qwen3-0.6B, Training Step=10, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 11.1 | |
| BaseStudent Model=Qwen3-0.6B, Teacher Model=Qwen3-4B-Instruct-25072026.06 | 1.5 |