Daquan Zhou

Daquan Zhou is a Tenure-track Assistant Professor at Peking University, jointly appointed in the School of AI for Science and the School of Electrical and Computer Engineering. He earned his PhD from the National University of Singapore and previously conducted research at NVIDIA Research and ByteDance. He is best known for his work on efficient video generation models, including HunyuanVideo, MagicVideo, and StoryDiffusion, as well as contributions to efficient network architecture design such as Coordinate Attention and DeepViT.