Fetching the paper…

MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation? · Around