Mathematical Reasoning on HMMT25 (Avg@16)
25Avg@16DASD
Evaluation Results
| Method | Links | |
|---|---|---|
| DASDModel Scale=4B2026.05 | 25 | |
| DASDModel Scale=8B2026.05 | 25 | |
| GRPOModel Scale=4B2026.05 | 22.1 | |
| HEPOModel Scale=4B2026.05 | 21.2 | |
| GRPOModel Scale=8B2026.05 | 20.8 | |
| RLSDModel Scale=8B2026.05 | 20.8 | |
| RLSDModel Scale=4B2026.05 | 20.4 | |
| HEPOModel Scale=8B2026.05 | 20.4 | |
| DASDModel Scale=1.7B2026.05 | 17.7 | |
| SRPOModel Scale=4B2026.05 | 17.7 | |
| SRPOModel Scale=8B2026.05 | 17.5 | |
| HEPOModel Scale=1.7B2026.05 | 15.6 | |
| RLSDModel Scale=1.7B2026.05 | 15.2 | |
| Qwen3Model Scale=4B2026.05 | 15.2 | |
| OPSDModel Scale=8B2026.05 | 15 | |
| SDPOModel Scale=8B2026.05 | 14.4 | |
| SDPOModel Scale=4B2026.05 | 13.3 | |
| Qwen3Model Scale=8B2026.05 | 13.1 | |
| OPSDModel Scale=4B2026.05 | 12.9 | |
| GRPOModel Scale=1.7B2026.05 | 11.5 | |
| SRPOModel Scale=1.7B2026.05 | 10.2 | |
| OPSDModel Scale=1.7B2026.05 | 8.5 | |
| SDPOModel Scale=1.7B2026.05 | 8.1 | |
| Qwen3Model Scale=1.7B2026.05 | 7.9 |