Sihan Xu

Vision-Language Models Are Not Pragmatically Competent in Referring Expression Generation

4D-LRM: Large Space-Time Reconstruction Model From and To Any View at Any Time

Multi-Object Hallucination in Vision Language Models