Mathematical Reasoning on AIME 2025 (Weighted majority-voting accuracy, Pass@128, Pass@1)
69.32Weighted Majority Voting AccuracyOTV
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| OTVBackbone=QWEN3-4B2026.03 | 69.32 | — | — | |
| DeepConfBackbone=QWEN3-4B2026.03 | 66.77 | — | — | |
| GenRMBackbone=QWEN3-4B2026.03 | 66.72 | — | — | |
| VersaPRM-8BBackbone=QWEN3-4B2026.03 | 66.72 | — | — | |
| Math-Shepherd-7BBackbone=QWEN3-4B2026.03 | 66.56 | — | — | |
| MAJ@128Backbone=QWEN3-4B2026.03 | 66.46 | — | — | |
| Qwen2.5-PRM800K-7BBackbone=QWEN3-4B2026.03 | 66.35 | — | — | |
| Qwen2.5-PRM-7BBackbone=QWEN3-4B2026.03 | 63.49 | — | — | |
| AceMath-RM-7BBackbone=QWEN3-4B2026.03 | 60 | — | — | |
| OTVBackbone=DAPO-QWEN-32B2026.03 | 49.58 | — | — | |
| Qwen2.5-PRM800K-7BBackbone=DAPO-QWEN-32B2026.03 | 48.18 | — | — | |
| Qwen2.5-PRM-7BBackbone=DAPO-QWEN-32B2026.03 | 46.77 | — | — | |
| DeepConfBackbone=DAPO-QWEN-32B2026.03 | 44.79 | — | — | |
| Math-Shepherd-7BBackbone=DAPO-QWEN-32B2026.03 | 43.65 | — | — | |
| VersaPRM-8BBackbone=DAPO-QWEN-32B2026.03 | 43.18 | — | — | |
| AceMath-RM-7BBackbone=DAPO-QWEN-32B2026.03 | 42.76 | — | — | |
| GenRMBackbone=DAPO-QWEN-32B2026.03 | 42.55 | — | — | |
| MAJ@128Backbone=DAPO-QWEN-32B2026.03 | 41.77 | — | — | |
| OTVBackbone=QWEN3-8B2026.03 | 33.85 | — | — | |
| GenRMBackbone=QWEN3-8B2026.03 | 32.66 | — | — | |
| DeepConfBackbone=QWEN3-8B2026.03 | 31.87 | — | — | |
| Qwen2.5-PRM-7BBackbone=QWEN3-8B2026.03 | 29.48 | — | — | |
| Qwen2.5-PRM800K-7BBackbone=QWEN3-8B2026.03 | 29.32 | — | — | |
| VersaPRM-8BBackbone=QWEN3-8B2026.03 | 29.11 | — | — | |
| MAJ@128Backbone=QWEN3-8B2026.03 | 28.44 | — | — | |
| AceMath-RM-7BBackbone=QWEN3-8B2026.03 | 26.72 | — | — | |
| Math-Shepherd-7BBackbone=QWEN3-8B2026.03 | 26.04 | — | — | |
| Pass@1Backbone=QWEN3-4B2026.03 | — | — | 46.67 | |
| Pass@1Backbone=QWEN3-8B2026.03 | — | — | 19.32 | |
| Pass@1Backbone=DAPO-QWEN-32B2026.03 | — | — | 36.42 | |
| Pass@128Backbone=QWEN3-4B2026.03 | — | 83.13 | — | |
| Pass@128Backbone=QWEN3-8B2026.03 | — | 56.98 | — | |
| Pass@128Backbone=DAPO-QWEN-32B2026.03 | — | 68.75 | — |