Mathematical Reasoning on AMC 2023 (test) (Avg@8)
68.7Avg@8 Success RateDIFF
Evaluation Results
| Method | Links | |
|---|---|---|
| DIFFModel=Qwen2.5-Math-7B, Evaluation Step=9002026.05 | 68.7 | |
| GAINRLModel=Qwen2.5-Math-7B, Evaluation Step=9002026.05 | 68.4 | |
| SAERLGModel=Qwen2.5-Math-7B, Evaluation Step=9002026.05 | 68.4 | |
| GRPOModel=Qwen2.5-Math-7B, Evaluation Step=9002026.05 | 68.1 | |
| ADARFTModel=Qwen2.5-Math-7B, Evaluation Step=9002026.05 | 63.1 | |
| SAERLGModel=Qwen2.5-Math-1.5B, Evaluation Step=9002026.05 | 56.2 | |
| SAERLDModel=Qwen2.5-Math-1.5B, Evaluation Step=9002026.05 | 55.9 | |
| DAPOModel=Qwen2.5-Math-1.5B, Evaluation Step=9002026.05 | 55.6 | |
| ADARFTModel=Qwen2.5-Math-1.5B, Evaluation Step=9002026.05 | 55.6 | |
| DIFFModel=Qwen2.5-Math-1.5B, Evaluation Step=9002026.05 | 55 | |
| GRPOModel=Qwen2.5-Math-1.5B, Evaluation Step=9002026.05 | 53.4 | |
| GAINRLModel=Qwen2.5-Math-1.5B, Evaluation Step=9002026.05 | 53.1 |