| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Process Reward Modeling | PRMBENCH | Simplicity Avg Score81.7 | 22 | |
| Step-level reasoning evaluation | PRMBench step-level 2025 | FPR47.13 | 5 | |
| Step-level quality assessment | PRMBench | Simplicity54.65 | 5 | |
| Mathematical Reasoning | PRMBench | PRMScore60.7 | 4 |