Mathematical Reasoning on AIME 2025 (p@1, p@16)
82.08P@1DoLa
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DoLaModel=Qwen3-30B-A3B-Thinking2026.02 | 82.08 | 93.33 | |
| CoTModel=Qwen3-30B-A3B-Thinking2026.02 | 81.25 | 90 | |
| LEDModel=Qwen3-30B-A3B-Thinking2026.02 | 81.04 | 93.33 | |
| ST-GModel=Qwen3-30B-A3B-Thinking2026.02 | 79.79 | 93.33 | |
| LEDModel=Qwen3-4B-Thinking2026.02 | 76.46 | 90 | |
| STModel=Qwen3-30B-A3B-Thinking2026.02 | 75.21 | 90 | |
| CoTModel=Qwen3-4B-Thinking2026.02 | 74.17 | 90 | |
| STModel=Qwen3-4B-Thinking2026.02 | 72.29 | 83.33 | |
| DoLaModel=Qwen3-4B-Thinking2026.02 | 72.08 | 86.67 | |
| ST-GModel=Qwen3-4B-Thinking2026.02 | 67.5 | 90 | |
| LEDModel=MiMo-7B-RL2026.02 | 60.62 | 83.33 | |
| CoTModel=MiMo-7B-RL2026.02 | 59.17 | 80 | |
| DoLaModel=MiMo-7B-RL2026.02 | 59.17 | 80 | |
| ST-GModel=MiMo-7B-RL2026.02 | 59.17 | 80 | |
| STModel=MiMo-7B-RL2026.02 | 54.79 | 80 |