Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

PRMBENCH

Benchmarks

Task NameDataset NameSOTA ResultTrend
Process Reward ModelingPRMBENCH
Simplicity Avg Score81.7
22
Step-level quality assessmentPRMBench
Simplicity54.65
5
Mathematical ReasoningPRMBench
PRMScore60.7
4
Showing 3 of 3 rows