Step-level verification on ProcessBench (test)
82.1MATH AccuracyVeriBound
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| VeriBoundVerification Mode=Empirical2026.06 | 82.1 | 85.6 | 74.8 | 72 | 78.6 | |
| FOVER2026.06 | 81.4 | 84.9 | 74.1 | 71.2 | 77.9 | |
| ReasonFlux-PRM2026.06 | 80.1 | 83.7 | 72.5 | 69.8 | 76.5 | |
| VeriBoundVerification Mode=Bound (Theoretical)2026.06 | 79.8 | 83.2 | 72.6 | 69.5 | 76.3 | |
| R-PRM2026.06 | 79.3 | 82.4 | 71.8 | 68.9 | 75.6 | |
| PRM800K2026.06 | 78.1 | 81.6 | 70.2 | 67.5 | 74.4 | |
| Math-Shepherd2026.06 | 74.5 | 78.3 | 66.8 | 63.7 | 70.8 | |
| ORM2026.06 | 68.2 | 72.1 | 61.4 | 58.3 | 65 |