Scientific Reasoning on phys
54.5Pass RateSRPO
Evaluation results
| Method | Links | |
|---|---|---|
| SRPOModel=OLMo-3-7B-Instruct2026.05 | 54.5 | |
| SPO-TreeModel=OLMo-3-7B-Instruct2026.05 | 47.1 | |
| Cr-GRPOModel=OLMo-3-7B-Instruct2026.05 | 46.9 | |
| SRPOModel=Qwen2.5-14B-Instruct2026.05 | 45.5 | |
| BaseModel=OLMo-3-7B-Instruct2026.05 | 45.4 | |
| GRPOModel=OLMo-3-7B-Instruct2026.05 | 42.3 | |
| RRPOModel=Qwen2.5-14B-Instruct2026.05 | 39.8 | |
| RRPOModel=OLMo-3-7B-Instruct2026.05 | 33.9 | |
| SCoReModel=Qwen2.5-14B-Instruct2026.05 | 29.5 | |
| SCoReModel=OLMo-3-7B-Instruct2026.05 | 29.4 | |
| GRPOModel=Qwen2.5-14B-Instruct2026.05 | 26.9 | |
| BaseModel=Qwen2.5-14B-Instruct2026.05 | 26.6 | |
| Cr-GRPOModel=Qwen2.5-14B-Instruct2026.05 | 24.9 | |
| SPO-TreeModel=Qwen2.5-14B-Instruct2026.05 | 24.5 |