Mathematical Reasoning on AIME 2024 & 2025 (AUROC)
94.1AUROCSCVC
Evaluation Results
| Method | Links | |
|---|---|---|
| SCVCSampling budget (K)=82026.03 | 94.1 | |
| SCVCSampling budget (K)=52026.03 | 91.9 | |
| SCSampling budget (K)=82026.03 | 90 | |
| SCVCSampling budget (K)=22026.03 | 89.5 | |
| SCSampling budget (K)=52026.03 | 85.9 | |
| SCSampling budget (K)=22026.03 | 82.6 | |
| VCSampling budget (K)=82026.03 | 81.7 | |
| VCSampling budget (K)=52026.03 | 78.8 | |
| VCSampling budget (K)=22026.03 | 67.2 | |
| VCSampling budget (K)=12026.03 | 66.4 |