Han Yang (杨涵)

photo.jpg

I am an incoming PhD student at University of Massachusetts Amherst (UMass Amherst), where I’m fortunate to be advised by Prof. Chuang Gan. I recently graduated from The Chinese University of Hong Kong (CUHK) with a first-honors degree in Computer Science. During my undergraduate studies, I conducted research in embodied AI, multimodal learning, and generative models under the supervision of Prof. Chuang Gan at UMass Amherst and Prof. Yilun Du at Harvard. I also worked with Prof. Chi-Wing Fu at CUHK, focusing on low-level vision and self/semi-supervised learning.

My research interests lie in embodied AI and robot learning. My goal is to build intelligent agents that can navigate and perform tasks in human environments with accuracy and robustness. I’m particularly interested in robotic perception—how agents can actively perceive, explore, and build representations of their surroundings—and interaction with the physical world, which involves both high-level decision-making and robust low-level motion planning for complex tasks.

Publications

  1. CVPR’25
    preview_snapmem.gif
    3D-Mem: 3D Scene Memory for Embodied Exploration and Reasoning
    CVPR, 2025
  2. AAAI’25
    preview_unimumo.gif
    UniMuMo: Unified Text, Music, and Motion Generation
    Han YangKun Su, Yutong Zhang, Jiaben Chen, Kaizhi Qian, Gaowen Liu, and Chuang Gan
    AAAI, 2025
  3. ICCV’23
    preview_silt.png
    SILT: Shadow-aware Iterative Label Tuning for Learning to Detect Shadows from Noisy Labels
    Han Yang*Tianyu Wang*, Xiaowei Hu, and Chi-Wing Fu
    ICCV, 2023