| Benchmark | Model | Rank | Results |
|---|---|---|---|
| mathematical-reasoning-on-aime-2025 | DeepSeek-R1-Distill-Qwen-7B (Graph-based CoT Pruning) | – | Accuracy: 31.67 |
| mathematical-reasoning-on-math-500 | DeepSeek-R1-Distill-Qwen-1.5B (Graph-based CoT Pruning) | – | Accuracy: 80.4 |
| mathematical-reasoning-on-olympiadbench | DeepSeek-R1-Distill-Qwen-7B (Graph-based CoT Pruning) | – | Accuracy: 59.85 |