Mathematical Reasoning on HMMT-25 (mean@8, pass@8)
50.83Mean@8DeepSeek-R1-Distill-Qwen-32B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DeepSeek-R1-Distill-Qwen-32B2026.05 | 50.83 | 70 | |
| SkyWork-OR1-Math-7B2026.05 | 32.22 | 55.17 | |
| GRPOStudent Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.05 | 32.22 | 55.17 | |
| LGRStudent Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.05 | 31.67 | 60 | |
| REOPOLDStudent Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.05 | 30 | 56.67 | |
| OPD (topk)Student Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.05 | 29.58 | 56.67 | |
| LGR (topk)Student Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.05 | 27.92 | 53.33 | |
| OPD (9k)Student Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.05 | 27.5 | 50 | |
| JSDStudent Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.05 | 25.83 | 43.33 | |
| Base ModelStudent Model=DeepSeek-R1-Distill-Qwen-7B2026.05 | 24.17 | 43.33 | |
| LGRStudent Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.05 | 22.08 | 36.67 | |
| LGR (topk)Student Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.05 | 22.08 | 46.67 | |
| OPD (topk)Student Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.05 | 20.5 | 27.59 | |
| OPD (9k)Student Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.05 | 20.42 | 36.67 | |
| REOPOLDStudent Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.05 | 19.17 | 26.67 | |
| JSDStudent Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.05 | 18.75 | 23.33 | |
| GRPOStudent Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.05 | 16.74 | 20.69 | |
| Base ModelStudent Model=DeepSeek-R1-Distill-Qwen-1.5B2026.05 | 11.67 | 23.33 |