List your company

Image Generation Assessment on MMRB2 (test)

69.2AccuracyUnifiedReward-Flex
54.43258.26662.165.934Feb 2, 2026

Evaluation results

MethodLinks
UnifiedReward-Flex2026.0269.2
UnifiedReward-FlexDPO=False, Protocol=Both correct2026.0268.4
UnifiedReward-FlexDPO=False2026.0267.5
UnifiedReward-Think2026.0266
UnifiedReward2026.0260
HPSv32026.0258.5
PickScore2026.0257.6
HPSv22026.0255