Diffusion Models are Zero-Shot Generative Text-Vision Retrievers.

Bao Li, Zeke Xie, Xiaomei Zhang, Xiangyu Zhu, Zhen Lei

Browse the full ICASSP paper archive.