Mathematical Reasoning on MATH 500 (Acc, Len)
89AccuracyCyclicReflex
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CyclicReflexBase Model=DeepSeek-R1-Distill-Qwen-7B2025.06 | 89 | 777.93 | |
| TIPBase Model=DeepSeek-R1-Distill-Qwen-7B2025.06 | 87 | 775.77 | |
| OriginalBase Model=DeepSeek-R1-Distill-Qwen-7B2025.06 | 86 | 785.25 | |
| CyclicReflexBase Model=DeepSeek-R1-Distill-Llama-8B2025.06 | 85 | 1,108.3 | |
| S1Base Model=DeepSeek-R1-Distill-Qwen-7B2025.06 | 83 | 1,190.96 | |
| OriginalBase Model=DeepSeek-R1-Distill-Llama-8B2025.06 | 83 | 1,196.98 | |
| TIPBase Model=DeepSeek-R1-Distill-Llama-8B2025.06 | 83 | 1,080.62 | |
| S1Base Model=DeepSeek-R1-Distill-Llama-8B2025.06 | 78 | 1,461.93 | |
| CyclicReflexBase Model=DeepSeek-R1-Distill-Qwen-1.5B2025.06 | 77 | 1,212.94 | |
| TIPBase Model=DeepSeek-R1-Distill-Qwen-1.5B2025.06 | 75 | 1,206.91 | |
| OriginalBase Model=DeepSeek-R1-Distill-Qwen-1.5B2025.06 | 74 | 1,253.05 | |
| S1Base Model=DeepSeek-R1-Distill-Qwen-1.5B2025.06 | 73 | 1,532.05 |