VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generation

Publication image