Fetching the paper…

Image Anything: Towards Reasoning-coherent and Training-free Multi-modal Image Generation · Around