List your company

Math Reasoning on AIME 2024 (Pass@1, Pass@16)

41.66Pass@1EAPO
31.270433.967736.66539.3623Jun 29, 2026

Evaluation results

MethodLinks
EAPOModel=Qwen-3-8B-Base2026.0641.6674.37
EAPOModel=Qwen-2.5-Math-7B2026.0640.1460.44
EAPO w/o sISModel=Qwen-2.5-Math-7B2026.0639.6863.29
MOPDModel=Qwen-2.5-Math-7B2026.0638.8560.28
OPDModel=Qwen-3-8B-Base2026.0638.2367.48
OPDModel=Qwen-2.5-Math-7B2026.0638.1255.06
MOPDModel=Qwen-3-8B-Base2026.0637.874.68
Trajectory ReplayModel=Qwen-2.5-Math-7B2026.0637.2558.6
DAPOModel=Qwen-2.5-Math-7B2026.0637.1958
DAPOModel=Qwen-3-8B-Base2026.0636.6771.41
GRPOModel=Qwen-2.5-Math-7B2026.0632.0850.24
GRPOModel=Qwen-3-8B-Base2026.0631.6761.98