Jiaming Song is a research scientist at NVIDIA Research. Prior to joining NVIDIA, he did his Ph.D. at Stanford University, under the supervision of Stefano Ermon. He is the creator of DDIM, the earliest accelerated algorithm for diffusion models that is widely used in recent generative AI systems including DALL-E 2, Imagen, Stable Diffusion, and ERNIE-ViLG 2.0. He also served a critical role in developing SDEdit, which is used in Stable Diffusion for image-to-image translation. He also co-developed eDiff-I, NVIDIA's first large-scale text-to-image diffusion model. He is a recipient of the ICLR 2022 Outstanding Paper Award.