MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation? Paper • 2407.04842 • Published Jul 5 • 52 • 5