Mathematical Reasoning on AMC 23 (Pass@8, Pass@16, Pass@32)
92.5Pass@8w/ Clip
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| w/ ClipBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.12 | 92.5 | 92.5 | 90 | |
| w/ CovBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.12 | 92.5 | 90 | 97.5 | |
| SENTBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.12 | 92.5 | 95 | 97.5 | |
| w/ EnBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.12 | 90 | 90 | 97.5 | |
| w/ AdvBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.12 | 90 | 92.5 | 95 | |
| w/ MaskBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.12 | 90 | 92.5 | 95 | |
| GRPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.12 | 87.5 | 95 | 95 | |
| w/ High EnBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.12 | 85 | 92.5 | 95 | |
| Base modelBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.12 | 80 | 87.5 | 92.5 |