Mathematical Reasoning on MathMCQA
59.07AccuracySPEAR
Evaluation Results
| Method | Links | |
|---|---|---|
| SPEARModel=Qwen2.5-1.5B2026.05 | 59.07 | |
| SPEARModel=Llama3.2-3B2026.05 | 56.7 | |
| Feedback SFTModel=Qwen2.5-1.5B2026.05 | 54.93 | |
| GRPOModel=Qwen2.5-1.5B2026.05 | 50.87 | |
| Feedback SFTModel=Llama3.2-3B2026.05 | 48.6 | |
| RLTF-SDModel=Llama3.2-3B2026.05 | 44.7 | |
| RLTF-SDModel=Qwen2.5-1.5B2026.05 | 44.7 | |
| GRPOModel=Llama3.2-3B2026.05 | 40.2 | |
| OPSDModel=Llama3.2-3B2026.05 | 0 | |
| OPSDModel=Qwen2.5-1.5B2026.05 | 0 |