Mathematical Reasoning on Math (Pass Rate)
55.9Pass RateSRPO
Evaluation results
| Method | Links | |
|---|---|---|
| SRPOModel=OLMo-3-7B-Instruct2026.05 | 55.9 | |
| GRPOModel=OLMo-3-7B-Instruct2026.05 | 51.3 | |
| Cr-GRPOModel=OLMo-3-7B-Instruct2026.05 | 47 | |
| SPO-TreeModel=OLMo-3-7B-Instruct2026.05 | 47 | |
| BaseModel=OLMo-3-7B-Instruct2026.05 | 44.6 | |
| SRPOModel=Qwen2.5-14B-Instruct2026.05 | 39.3 | |
| RRPOModel=OLMo-3-7B-Instruct2026.05 | 37.9 | |
| RRPOModel=Qwen2.5-14B-Instruct2026.05 | 37.5 | |
| SCoReModel=OLMo-3-7B-Instruct2026.05 | 32.6 | |
| SCoReModel=Qwen2.5-14B-Instruct2026.05 | 30.9 | |
| BaseModel=Qwen2.5-14B-Instruct2026.05 | 30 | |
| Cr-GRPOModel=Qwen2.5-14B-Instruct2026.05 | 29.8 | |
| GRPOModel=Qwen2.5-14B-Instruct2026.05 | 29.4 | |
| SPO-TreeModel=Qwen2.5-14B-Instruct2026.05 | 29 |