AIME-26

AIME-26 is a benchmark dataset used to evaluate reinforcement learning with verifiable rewards (RLVR) in the context of solving challenging math questions.

Papers using AIME-26 (2)

Weight-Space Geometry of Offline Reasoning Training2026Mitigating Distribution Sharpening in Math RLVR via Distribution-Aligned Hint Synthesis and Backward Hint Annealing2026