Mathematical Reasoning on MinervaMath (test)
47.79Pass@1GRPO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GRPOStudent=Qwen2.5 14B2025.05 | 47.79 | — | |
| Qwen2.5-14B (R)Role=Standalone Teacher2025.05 | 47.79 | — | |
| GRPOStudent=Qwen2.5 32B2025.05 | 46.69 | — | |
| W2SRStudent=Qwen2.5 32B, Teacher=Qwen2.5-14B (R)2025.05 | 45.59 | 200 | |
| W2SR-PStudent=Qwen2.5 14B, Teacher=Qwen2.5-7B (R)2025.05 | 45.22 | 70.86 | |
| W2SR-PStudent=Qwen2.5 32B, Teacher=Qwen2.5-14B (R)2025.05 | 44.85 | 267.27 | |
| W2SRStudent=Qwen2.5 32B, Teacher=Qwen2.5-7B (R)2025.05 | 42.28 | 42.88 | |
| W2SRStudent=Qwen2.5 14B, Teacher=Qwen2.5-7B (R)2025.05 | 41.54 | 29.14 | |
| W2SR-PStudent=Qwen2.5 32B, Teacher=Qwen2.5-7B (R)2025.05 | 40.81 | 23.83 | |
| W2SR-PStudent=Qwen2.5 32B, Teacher=Qwen2.5-1.5B (R)2025.05 | 40.07 | 73.91 | |
| W2SR-PStudent=Qwen2.5 Math-7B, Teacher=Qwen2.5-1.5B (R)2025.05 | 39.71 | 108.69 | |
| Qwen2.5-7B (R)Role=Standalone Teacher2025.05 | 38.97 | — | |
| GRPOStudent=Qwen2.5 Math-7B2025.05 | 38.24 | — | |
| W2SR-PStudent=Qwen2.5 14B, Teacher=Qwen2.5-1.5B (R)2025.05 | 38.24 | 63.92 | |
| W2SRStudent=Qwen2.5 Math-7B, Teacher=Qwen2.5-1.5B (R)2025.05 | 37.5 | 95.63 | |
| W2SRStudent=Qwen2.5 32B, Teacher=Qwen2.5-1.5B (R)2025.05 | 37.13 | 62.32 | |
| W2SR-PStudent=Qwen2.5 Math-7B, Teacher=Qwen2.5-0.5B (R)2025.05 | 36.4 | 93.5 | |
| W2SRStudent=Qwen2.5 14B, Teacher=Qwen2.5-1.5B (R)2025.05 | 32.72 | 43.07 | |
| LIMOStudent=Qwen2.5 14B2025.05 | 31.25 | — | |
| LIMOStudent=Qwen2.5 Math-7B2025.05 | 30.88 | — | |
| LIMOStudent=Qwen2.5 32B2025.05 | 30.51 | — | |
| VanillaStudent=Qwen2.5 32B2025.05 | 29.04 | — | |
| VanillaStudent=Qwen2.5 14B2025.05 | 28.31 | — | |
| W2SRStudent=Qwen2.5 Math-7B, Teacher=Qwen2.5-0.5B (R)2025.05 | 26.84 | 59.73 | |
| Qwen2.5-1.5B (R)Role=Standalone Teacher2025.05 | 21.32 | — | |
| VanillaStudent=Qwen2.5 Math-7B2025.05 | 13.6 | — | |
| Qwen2.5-0.5B (R)Role=Standalone Teacher2025.05 | 9.93 | — |