Scientific Reasoning on chem
26.5Pass RateGRPO
Evaluation results
| Method | Links | |
|---|---|---|
| GRPOModel=Qwen2.5-14B-Instruct2026.05 | 26.5 | |
| RRPOModel=Qwen2.5-14B-Instruct2026.05 | 25.6 | |
| SCoReModel=Qwen2.5-14B-Instruct2026.05 | 25 | |
| SRPOModel=OLMo-3-7B-Instruct2026.05 | 24.7 | |
| SRPOModel=Qwen2.5-14B-Instruct2026.05 | 24.5 | |
| BaseModel=Qwen2.5-14B-Instruct2026.05 | 23.6 | |
| SPO-TreeModel=Qwen2.5-14B-Instruct2026.05 | 23.4 | |
| Cr-GRPOModel=Qwen2.5-14B-Instruct2026.05 | 22.9 | |
| Cr-GRPOModel=OLMo-3-7B-Instruct2026.05 | 21.4 | |
| SPO-TreeModel=OLMo-3-7B-Instruct2026.05 | 20.9 | |
| GRPOModel=OLMo-3-7B-Instruct2026.05 | 20.7 | |
| RRPOModel=OLMo-3-7B-Instruct2026.05 | 20.5 | |
| BaseModel=OLMo-3-7B-Instruct2026.05 | 19.8 | |
| SCoReModel=OLMo-3-7B-Instruct2026.05 | 18.9 |