Mathematical Reasoning on AMC23 (Avg@32, Pass@32)
91.4Average Score @32MASPO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MASPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.02 | 91.4 | 95 | |
| BAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.02 | 89.2 | 97.2 | |
| Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-7B2026.02 | 89 | 96.6 | |
| SAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.02 | 88.7 | 95 | |
| GRPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.02 | 88.1 | 96.6 | |
| DACBackbone=DeepSeek-R1-Distill-Qwen-7B2026.02 | 88.1 | 99.6 | |
| Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-7B2026.02 | 87.7 | 96.6 | |
| Clip HigherBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.02 | 82.8 | 94.9 | |
| SAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.02 | 82.7 | 94.8 | |
| MASPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.02 | 82.2 | 95 | |
| DACBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.02 | 80.2 | 95 | |
| BAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.02 | 80.1 | 94.9 | |
| GRPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.02 | 79.5 | 94.8 | |
| Entropy Adv.Backbone=DeepSeek-R1-Distill-Qwen-1.5B2026.02 | 77.1 | 94.1 |