List your company

Mathematical Reasoning on AMC 23 (Pass@8)

74.53Pass@8POPO
33.647644.261354.87565.4887May 7, 2026

Evaluation results

MethodLinks
POPOModel=DeepSeek-Math-7B2026.0574.53
GRPOModel=DeepSeek-Math-7B2026.0573.38
POPOModel=Llama-3.1-8B2026.0537.78
GRPOModel=Llama-3.1-8B2026.0535.22