Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

PRM800K dataset and benchmarks · SOTA2 Research

Research Datasets

PRM800K

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Mathematical Reasoning

PRM800K (test)

80Accuracy

15

First-error detection

PRM800K

92.9Accuracy

6

Step-level hallucination detection

PRM800K

99.8AUROC

6

Process Reward Modeling

PRM800K (test)

95.5Pos-Acc

5

Stepwise Confidence Attribution

PRM800K

0.818AUROC

5

PRM800K

0.613AUC-ROC

5

Instance-level Evaluation

PRM800K

0.42AUC-ROC

1

Popular tasks

Mathematical Reasoning Math Reasoning Process Reward Modeling Instance-level Evaluation Step-level hallucination detection

Follow for updates

Join SOTA2 Contact