List your company

Mathematical Reasoning on AMC23 (Last/Best Score)

94.92Last ScoreRPG-UFKL
90.936891.970993.00594.0391May 23, 2025

Evaluation results

MethodLinks
RPG-UFKLContext length=8k2025.0594.9295.17
RPG-REINFORCE-UFKLContext length=8k2025.0594.5394.69
RPG-URKLContext length=8k2025.0594.0695.39
RPG-REINFORCE-URKLContext length=8k2025.0593.9894.69
DAPOContext length=8k2025.0593.7594.3
GRPOContext length=8k2025.0591.0992.97