Jinzhou Li

I'm an incoming Robotics Ph.D. student at Duke University. Currently, I'm a Visiting Research Student at the Center on Frontiers of Computing Studies (CFCS) in the School of Computer Science at Peking University, supervised by Prof. Hao Dong. Additionally, I am a research intern at AGI-Bot.

Previously, I received my Master's degree from Cornell University, where I was advised by Prof. Maha Haji and worked with Prof. Daniel Hastings from MIT. Prior to this, I studied Computer Science at the University of Vermont.

Happy to chat about research!

profile photo

Publications

My research focuses on enabling robots to achieve human-level dexterity in complex environments. I work on bridging the gap between human and robotic capabilities through dexterous manipulation, tactile sensing, and machine learning approaches.

* Equal Contribution

Adaptive Visual-Tacile Fusion with Predictive Force Attention for Dexterous Manipulation

Jinzhou Li*, Tianhao Wu*, Jiyao Zhang**, Zeyuan Chen**, Haotian Jin, Mingdong Wu, Yujun Shen, Yaodong Yang, Hao Dong
website, code

A future force-guided attention fusion module that adaptively adjusts the weights of visual and tactile features.



Canonical Representation and Force-Based Pretraining of 3D Tactile for Dexterous Visuo-Tactile Policy Learning

Tianhao Wu, Jinzhou Li*, Jiyao Zhang*, Mingdong Wu, Hao Dong
IEEE International Conference on Robotics and Automation (ICRA), 2025
arXiv, website, code

A novel 3D tactile data representation and force-based pretraining to enhance dexterous manipulation learning.



HGIC: A Hand Gesture-Based Interactive Control System for Efficient and Scalable Multi-UAV Operations

Jinzhou Li, Mengsha Hu, Lei Xu, Yibei Guo, Rui Liu
IEEE RO-MAN 2024, Paper
IEEE MRS 2023, Poster Presentation
arXiv

Hand gesture-based interactive control system for efficient and scalable multi-UAV operations.

Teaching

  • Teaching Assistant - Meta CS 4/5782: Intro to Deep Learning, Fall 2023, Cornell University

Professional Activities

  • Conference Reviewer: ICRA 2024, 2025


Oct, 2024.

Design and source code from Jon Barron's website.