List your company

Mathematical Reasoning on Olympiad (Mean@32 accuracy)

29.86Mean@32 AccuracyGRPO
21.352823.561425.7727.9786Jun 15, 2026

Evaluation results

MethodLinks
GRPOModel=Qwen2.5-Math-7B, G=22026.0629.86
GRPOModel=Qwen2.5-Math-7B, G=162026.0628.66
C-RF w/ NTFModel=Qwen2.5-Math-7B, G=12026.0626.54
RF++ w/ NTFModel=Qwen2.5-Math-7B, G=12026.0624.92
GRPOModel=Qwen2.5-Math-1.5B, G=162026.0624.86
RF++Model=Qwen2.5-Math-7B, G=82026.0624.74
C-RF w/ NTFModel=Qwen2.5-Math-1.5B, G=12026.0623.89
GRPOModel=Qwen2.5-Math-1.5B, G=22026.0623.52
RF++Model=Qwen2.5-Math-1.5B, G=82026.0622.87
RF++ w/ NTFModel=Qwen2.5-Math-1.5B, G=12026.0621.68