Reasoning on Average (MATH500, AMC23, AIME24, AIME25, GPQA-d)
87.2AccuracyDDC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DDCBackbone=Qwen3-32B, Voting size budget=5122026.05 | 87.2 | 4.9 | |
| D-LBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.05 | 87.1 | 16.4 | |
| DDCBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.05 | 86.7 | 5.7 | |
| D-LBackbone=Qwen3-32B, Voting size budget=5122026.05 | 86.1 | 19 | |
| SCBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.05 | 85.2 | 48.5 | |
| ACBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.05 | 85.2 | 17.1 | |
| D-HBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.05 | 85.2 | 28.5 | |
| D-HBackbone=Qwen3-32B, Voting size budget=5122026.05 | 85 | 23.2 | |
| SCBackbone=Qwen3-32B, Voting size budget=5122026.05 | 84.7 | 41.1 | |
| ACBackbone=Qwen3-32B, Voting size budget=5122026.05 | 84.4 | 13.7 | |
| DDCBackbone=Qwen3-4B, Voting size budget=5122026.05 | 83.4 | 4.9 | |
| DDCBackbone=Qwen3-8B, Voting size budget=5122026.05 | 81.4 | 5.1 | |
| D-LBackbone=Qwen3-8B, Voting size budget=5122026.05 | 81.3 | 15.8 | |
| D-HBackbone=Qwen3-8B, Voting size budget=5122026.05 | 81.3 | 24 | |
| SCBackbone=Qwen3-8B, Voting size budget=5122026.05 | 80.8 | 38.4 | |
| D-LBackbone=Qwen3-4B, Voting size budget=5122026.05 | 80.5 | 19.1 | |
| ACBackbone=Qwen3-8B, Voting size budget=5122026.05 | 79.1 | 13.1 | |
| D-HBackbone=Qwen3-4B, Voting size budget=5122026.05 | 75.8 | 28.3 | |
| SCBackbone=Qwen3-4B, Voting size budget=5122026.05 | 74.8 | 45 | |
| ACBackbone=Qwen3-4B, Voting size budget=5122026.05 | 74.8 | 14.9 | |
| DDCBackbone=Qwen3-1.7B, Voting size budget=5122026.05 | 70.1 | 8 | |
| D-LBackbone=Qwen3-1.7B, Voting size budget=5122026.05 | 69.4 | 36.3 | |
| D-HBackbone=Qwen3-1.7B, Voting size budget=5122026.05 | 69 | 50.4 | |
| SCBackbone=Qwen3-1.7B, Voting size budget=5122026.05 | 67 | 69.3 | |
| ACBackbone=Qwen3-1.7B, Voting size budget=5122026.05 | 66.9 | 28.4 |