Mathematical Reasoning on AMC 23 (Accuracy, Length)
86.9Accuracy (AMC 23)CES
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CESBackbone=DeepSeek-R1-Distill-Qwen-7B2026.05 | 86.9 | 3,160 | |
| DAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.05 | 85 | 4,174 | |
| Entropy ShapeBackbone=DeepSeek-R1-Distill-Qwen-7B2026.05 | 85 | 3,843 | |
| R1-7BBackbone=DeepSeek-R1-Distill-Qwen-7B2026.05 | 83.8 | 4,446 |