Gordon Qian is a research scientist at Snap Research Creative Vision team, leading the unified multimodal generation and its end-to-end R&D. He earned his Ph.D. in Computer Science from KAUST, where he was fortunate to be advised by Prof. Bernard Ghanem. He has authored 19 top-tier conference and journal papers and his publications have received over 3300 citations, and his current h-index is 19. His representative work includes PointNeXt (NeurIPS, >1100 cites, >900 GitHub stars), Magic123 (ICLR, >450 cites, >1.6K GitHub stars) and Omni-ID (CVPR'25). His leading authored papers SR-Training (NeurIPS), Omni-ID, and ComposeMe (Siggraph Asia) have been integrated into Snapchat products, serving 400 million monthly active users, with 6 filed patents. He also serves as area chair for ICLR 2026.
If you are interested in working in image/video generative models with me, please reach out at guocheng.qian [at] outlook.com