S. Sara MahdaviPhotorealistic Text-to-Image Diffusion Models with Deep Language UnderstandingAll names