MUMU: Bootstrapping Multimodal Image Generation from Text-to-Image Data Paper • 2406.18790 • Published 25 days ago • 32