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

Step-level quality assessment on PRMBench

54.65Simplicity

CPMI_Merge

26.5733.8641.1548.44Apr 12, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.04
54.6562.356.560.67
2026.04
52.5660.2155.5958.75
2026.04
47.1649.5647.249.64
2026.04
40.8836.9235.0638.8
2026.04
27.6521.6418.6523.2