Mathematical Reasoning on Average of Math Tasks
76.9AccuracyLaser-DE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Laser-DEBackbone=DeepSeek-Distill-Qwen-7B2025.09 | 76.9 | 2,554 | |
| Laser-DBackbone=DeepSeek-Distill-Qwen-7B2025.09 | 76.8 | 2,649 | |
| DAPOBackbone=DeepSeek-Distill-Qwen-7B2025.09 | 76 | 3,405 | |
| JETBackbone=DeepSeek-Distill-Qwen-7B2025.09 | 75.2 | 2,999 | |
| BaseBackbone=DeepSeek-Distill-Qwen-7B2025.09 | 74.5 | 4,026 | |
| SFTBackbone=DeepSeek-Distill-Qwen-7B2025.09 | 73.7 | 3,862 | |
| DPOBackbone=DeepSeek-Distill-Qwen-7B2025.09 | 73.4 | 3,839 | |
| AdaThinkBackbone=DeepSeek-Distill-Qwen-7B2025.09 | 73.2 | 2,720 | |
| LCR1Backbone=DeepSeek-Distill-Qwen-7B2025.09 | 71.8 | 2,238 | |
| JETBackbone=DS-Distill-Llama-8B2025.09 | 71.4 | 3,821 | |
| BaseBackbone=DS-Distill-Llama-8B2025.09 | 69.4 | 5,493 | |
| DAPOBackbone=DS-Distill-Llama-8B2025.09 | 67.8 | 4,081 | |
| Laser-DBackbone=DeepSeek-Distill-Qwen-1.5B2025.09 | 63.8 | 3,221 | |
| JETBackbone=DeepSeek-Distill-Qwen-1.5B2025.09 | 63.3 | 2,710 | |
| Laser-DEBackbone=DeepSeek-Distill-Qwen-1.5B2025.09 | 62.7 | 3,604 | |
| DPOBackbone=DS-Distill-Llama-8B2025.09 | 62.4 | 4,288 | |
| DAPOBackbone=DeepSeek-Distill-Qwen-1.5B2025.09 | 61.1 | 3,822 | |
| SFTBackbone=DS-Distill-Llama-8B2025.09 | 60.4 | 3,970 | |
| BaseBackbone=DeepSeek-Distill-Qwen-1.5B2025.09 | 58.9 | 4,765 | |
| AdaThinkBackbone=DeepSeek-Distill-Qwen-1.5B2025.09 | 58.8 | 2,948 | |
| SFTBackbone=DeepSeek-Distill-Qwen-1.5B2025.09 | 57.5 | 3,583 | |
| DPOBackbone=DeepSeek-Distill-Qwen-1.5B2025.09 | 57.2 | 3,744 | |
| LCR1Backbone=DeepSeek-Distill-Qwen-1.5B2025.09 | 54.4 | 2,682 |