Yuxiong Wang’s Homepage

Make things as simple as possible, but not simpler

– Albert Einstein

Welcome to my homepage. I am an Assistant Professor of Computer Science at the University of Illinois Urbana-Champaign (UIUC).

My research lies in computer vision, machine learning, and robotics, with a specific focus on open-world perception, meta-learning, multi-modal learning, generative modeling, and agent learning.

Before joining Illinois CS, I was a postdoctoral fellow in the Robotics Institute at Carnegie Mellon University, advised by Prof. Martial Hebert. I was a visitor to the Center for Data Science at New York University, working with Prof. Jean Ponce. I obtained my Ph.D. under the supervision of Prof. Martial Hebert in the Robotics Institute. I have also been closely working with Prof. Deva Ramanan and Prof. Ruslan Salakhutdinov. I receive the Best Paper Honorable Mention Award for streaming perception in ECCV 2020, and Best Paper Award Finalist in CVPR 2019, 2022. I am recognized as a Notable Area Chair in ICLR 2023, and an Expert Reviewer in TMLR. I am selected to participate in the National Academy of Engineering’s (NAE) Frontiers of Engineering symposium.

Please feel free to contact me if interested!

I am always seeking self-motivated Ph.D., M.S., and undergraduate students.
Our group also has openings for visiting students.

Current Research Topics

Meta-learning and learning to learn
Open-world, multi-modal, few-shot, and self-supervised learning
3D vision
Generative modeling, predictive learning
Human motion and human-object interaction modeling
Reinforcement learning, robot/agent learning
In-the-wild applications in robotics, autonomous driving, agriculture, materials science, chemistry, healthcare, etc.

Dissertation

Learning to Learn for Small Sample Visual Recognition

Thesis Committee:
Martial Hebert, Deva Ramanan, Ruslan Salakhutdinov, Andrew Zisserman, Yann LeCun

One of our primary research efforts focuses on bridging generative and discriminative learning, facilitating autonomous agents to perceive, interact, and act in the open world. For representative works, please refer to my two recent talks: the first talk, presented at the C3.ai Generative AI Workshop, elaborates on how we ground generative modeling in 3D and 4D; the second talk, given at the London Machine Learning Meetup, explores various techniques that leverage LLMs and generative visual models for perception and decision-making.