Computer vision
Teaching machines to interpret and reason about the visual world, from object recognition to medical imaging.
Publication
Why Latent Actions Fail, and How to Prevent It
Publication
LensErase: A Foundation Model for Erase Anything
Publication
Self-Evolving Visual Agent via Margin Reward
Career Opportunity
Research Intern – Document Intelligence and Content Understanding
Documents are one of the richest forms in which human knowledge is created and communicated, combining language, visual structure, tables, images, and domain-specific conventions. Our team is developing the next generation of document intelligence systems…