Yifan Hou

yifan_pic_new.jpg

yifhou at nvidia dot com

I am a senior research scientist at the NVIDIA GEAR Lab and an incoming Assistant Professor in the Department of Computer Science at the University of Texas, Austin. Before that, I was a Postdoctoral Scholar at Stanford University working with Prof. Shuran Song. I obtained my PhD from the Robotics Institute at Carnegie Mellon University, advised by Prof. Matthew T. Mason. I obtained BoE from the Department of Automation at Tsinghua University. I also spent time at Amazon Robotics, Toyota Research Institute and MIT.

My work focuses on extending the capabilities and robustness of robotic manipulation. I have been approaching the problem from the following perspectives:

  • Multi-modal imitation learning: Incorporating force/tactile as input modalities and using compliance control as an output modality. MCC, UMI-FT, Dex-UMI, ACP
  • DAgger/Post-training: Efficiently learning of contact-rich behavior from a small amount of rich data. MuSe, CR-DAgger
  • Sim2real transfer: Exploring physics safely in simulation to guide robots in the real world. DexMachina, SimPLE
  • Mechanics-based sythesis and optimization: Quantifying contact mode robustness and optimizing control for them. Hybrid Servoing, Shared Grasping, Planning

news

Jul 03, 2026 Checkout Multisensory Continual Learning (MuSe) for how we use a small amount of force data to unlock force-aware behavior on pretraining tasks.
May 26, 2026 I joined the NVIDIA GEAR Lab as a scientist.
May 05, 2026 I accepted the faculty offer from UT Austin CS and will start in Fall 2027.
Apr 27, 2026 Minimalist Compliance Control is accepted by RSS 2026.
Jan 31, 2026 UMI-FT is accepted by ICRA 2026.
Sep 29, 2025 DexUMI was selected as a best paper finalist at CoRL 2025!
Sep 27, 2025 CR-DAgger won the best paper award at the CoRL 2025 Human to Robot (H2R) workshop!
May 25, 2025 I gave an invited talk “Empower Robot Learning with Model-based Manipulation” at the Beyond Pick and Place workshop at ICRA 2025. Recording is available here.
May 21, 2025 Adaptive Compliance Policy won the best paper award at the ICRA 2025 Contact-rich Manipulation workshop.
Feb 19, 2025 I gave a lecture on “Introduction to Compliance Control” at Stanford EE/CS 381. Slides are available here.

selected publications

  1. umift.gif
    In-the-Wild Compliant Manipulation with UMI-FT
    Hojung Choi*, Yifan Hou*, Chuer Pan, Seongheon Hong, Austin Patel, Xiaomeng Xu, Mark R. Cutkosky, and Shuran Song
    2026
  2. cr-dagger.gif
    Compliant Residual DAgger: Improving Real-World Contact-Rich Manipulation with Human Corrections
    Xiaomeng Xu*, Yifan Hou*, Zeyi Liu, and Shuran Song
    2025
  3. dexmachina.gif
    DexMachina: Functional Retargeting for Bimanual Dexterous Manipulation
    Mandi Zhao, Yifan Hou, Dieter Fox, Yashraj Narang, Ajay Mandlekar, and Shuran Song
    2025
  4. dexumi.gif
    DexUMI: Using Human Hand as the Universal Manipulation Interface for Dexterous Manipulation
    Mengda Xu*, Han Zhang*, Yifan Hou, Zhenjia Xu, Linxi Fan, Manuela Veloso, and Shuran Song
    2025
  5. acp.gif
    Adaptive Compliance Policy: Learning Approximate Compliance for Diffusion Guided Control
    Yifan Hou, Zeyi Liu, Cheng Chi, Eric Cousineau, Naveen Kuppuswamy, Siyuan Feng, Benjamin Burchfiel, and Shuran Song
    In 2025 IEEE International Conference on Robotics and Automation (ICRA), 2025
  6. diamond.gif
    Manipulation with Shared Grasping
    Yifan Hou, Zhenzhong Jia, and Matthew T Mason
    Jun 2020
  7. robust_execution.gif
    Robust execution of contact-rich motion plans by hybrid force-velocity control
    Yifan Hou and Matthew T Mason
    In 2019 International Conference on Robotics and Automation (ICRA), Jun 2019