Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding Chitwan Saharia, William Chan