Mathematical Reasoning on OpenR1-Math-220k (unseen)
46AccuracyLMNet
Evaluation results
| Method | Links | |
|---|---|---|
| LMNetModel=Qwen2.5-1.5B, Training Data=7.47k+93.7k training2025.05 | 46 | |
| SFTModel=Qwen2.5-1.5B, Training Data=7.47k+93.7k training2025.05 | 34.7 | |
| PromptModel=Qwen2.5-1.5B2025.05 | 29 | |
| LMNetModel=Qwen2.5-0.5B, Training Data=7.47k+93.7k training2025.05 | 29 | |
| SFTModel=Qwen2.5-0.5B, Training Data=7.47k+93.7k training2025.05 | 23.2 | |
| PromptModel=Qwen2.5-0.5B2025.05 | 18 |