List your company

Multimodal Reasoning on MathVista (RC-Acc)

74.8RC-AccGroupwise Ranking Reward
66.06468.33270.672.868Apr 20, 2026

Evaluation results

MethodLinks
Groupwise Ranking RewardReward Method=Groupwise Ranking Reward, Training Budget=Matched, Training Epochs=22026.0474.8
Pointwise GRReward Method=Pointwise GR, Training Budget=Matched, Training Epochs=22026.0471
RLVRReference Checkpoint=RLVR2026.0470.8
Qwen2.5-VL-7B-ITReference Checkpoint=Qwen2.5-VL-7B-IT2026.0467.2
PRMReward Method=PRM, Training Budget=Matched, Training Epochs=22026.0466.4