Chen-Wei Xie

UniLiP: Adapting CLIP for Unified Multimodal Understanding, Generation and Editing

UFO: A Unified Approach to Fine-grained Visual Perception via Open-ended Language Interface

ACE: All-round Creator and Editor Following Instructions via Diffusion Transformer

Aligned Better, Listen Better for Audio-Visual Large Language Models

Asking Specifically Instead of Ambiguously to Your GPT Improves Image Caption