Xiao Yu Photo
Xiao Yu

I am currently a Ph.D. Student at Beijing Jiaotong University (BJTU) in MePro, co-supervised by Prof. Xiaojie Jin and Prof. Yunchao Wei.

My research interests primarily lie in Computer Vision and Machine Learning, with a specific focus on:

  • Vision World Models
  • Physics-grounded World Models
  • Multimodal Learning
Education
  • Ph.D. Student, Beijing Jiaotong University
    2026 - Present, Beijing, China
  • M.S. Student, Beijing Jiaotong University
    2024 - 2026, Beijing, China
  • B.E., Beijing Jiaotong University
    2020 - 2024, Beijing, China
Main Publications
Survey Thumbnail
From Seeing to Knowing the World: A Survey of Vision World Models
Xiao Yu*, Yichen Zhang, Mingzhang Wang, Shifang Zhao, Weizhe Liu, Yuyang Yin, Zhongwei Ren, Ning An, Xinglong Wu, Hao Liu, Houwen Peng, Yao Zhao, Jianchao Yang, Shuicheng Yan, Yunchao Wei, Xiaojie Jin*
Survey Paper

Topic: World Models

[Paper] [Project Page] [Awesome List]
Paper Thumbnail
PreFM: Online Audio-Visual Event Parsing via Predictive Future Modeling
Xiao Yu, Yan Fang, Xiaojie Jin, Yao Zhao, Yunchao Wei
Advances in Neural Information Processing Systems (NeurIPS), 2025

Topic: Multimodal Video Understanding

[Paper] [Code]
Paper Thumbnail
IPSeg: Image Posterior Mitigates Semantic Drift in Class-Incremental Segmentation
Xiao Yu*, Yan Fang*, Yao Zhao, Yunchao Wei
Arxiv, 2025

Topic: Continual Learning

[Paper] [Code]
Other Publications
Paper Thumbnail
VideoWorld 2: Learning Transferable Knowledge from Real-world Videos
Zhongwei Ren, Yunchao Wei, Xiao Yu, Guixun Luo, Yao Zhao, Bingyi Kang, Jiashi Feng, Xiaojie Jin
The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

Topic: World Models, Video Generation

[Paper] [Project] [Code]