Mathematical Reasoning on AIME 2024 (pass@K)
86.67Pass@KOriginal
Evaluation Results
| Method | Links | |
|---|---|---|
| OriginalBackbone=DSQwen-14B2025.10 | 86.67 | |
| STARBackbone=DSQwen-14B2025.10 | 86.67 | |
| SafeChainBackbone=DSQwen-14B2025.10 | 86.67 | |
| DAPOBackbone=DSQwen-14B2025.10 | 86.67 | |
| RECAPBackbone=DSQwen-14B2025.10 | 86.67 | |
| SFTBackbone=DSQwen-14B2025.10 | 83.33 | |
| OriginalBackbone=DSLlama-8B2025.10 | 70 | |
| SafeChainBackbone=DSLlama-8B2025.10 | 70 | |
| RECAPBackbone=DSLlama-8B2025.10 | 70 | |
| STARBackbone=DSLlama-8B2025.10 | 66.67 | |
| DAPOBackbone=DSLlama-8B2025.10 | 66.67 | |
| SFTBackbone=DSLlama-8B2025.10 | 63.33 |