Boyi Li

Latest

• Scaling Vision Pre-Training to 4K Resolution
• Efficient Video Diffusion Models via Content-Frame Motion-Latent Decomposition
• Geometry-informed neural operator for large-scale 3D PDEs