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

Math Reasoning on Mean of Six Math Benchmarks (Pass@K)

43.8Pass@1

GRPO

14.26421.93229.637.268Oct 21, 2025
Updated 11d ago

Evaluation Results

MethodLinks
2025.10
43.861.871.659.1
2025.10
39.656.367.554.5
2025.10
39.557.767.955
2025.10
39.256.666.154
2025.10
36.856.667.353.6
2025.10
36.656.86954.1
2025.10
32.354.665.350.7
2025.10
22.747.36143.7
2025.10
21.348.762.244.1
2025.10
18.237.75135.6
2025.10
15.935.550.333.9
2025.10
15.434.648.132.7