Yuanzhe Liu

Hi! 👋 I am an incoming Ph.D. student in Computer Science at the University of Illinois Urbana-Champaign. Previously, I received my M.S. in Data Science from the University of Pennsylvania and my B.S. in Mathematics from the University of Wisconsin-Madison.

My research focuses on generative models and reinforcement learning for embodied robotics, with the goal of enabling embodied agents to perceive, reason, and plan in the physical world. I hope to solve real-world challenges, transforming cutting-edge research into practical, scalable solutions.

Email / GitHub / Google Scholar

News

[2026/05] I've completed my master's degree at UPenn. Thank you to my advisors and friends for all your support!
[2026/02] PALM is accepted to CVPR 2026, my first robotics research paper. See you in Denver!
[2025/09] SpatialReasoner is accepted to NeurIPS 2025. Thanks to my collaborators at Google, the University of Illinois Urbana-Champaign, and Shanghai Jiao Tong University.

Selected Publications

--> For the full publication list, please refer to my Google Scholar.

All Selected Publications	3D Vision and Embodied Perception
Robust Robot Learning	Reinforcement Learning

PALM: Progress-Aware Policy Learning via Affordance Reasoning for Long-Horizon Robotic Manipulation

Yuanzhe Liu, Jingyuan Zhu, Yuchen Mo, Gen Li, Xu Cao, Jin Jin, Yifan Shen, Zhengyuan Li, Tianjiao Yu, Wenzhen Yuan, Fangqiang Ding, Ismini Lourentzou
CVPR, 2026
project page / arXiv

DeeptraceReward: Learning Human-Perceived Fakeness in Generated Videos with Multimodal LLMs

Xingyu Fu, Siyi Liu, Yinuo Xu, Pan Lu, Guangqiuse Hu, Tianbo Yang, Taran Anantasagar, Christopher Shen, Yikai Mao, Yuanzhe Liu, Keyush Shah, Chung Un Lee, Yejin Choi, James Zou, Dan Roth, Chris Callison-Burch
In Submission, 2026
project page / arXiv

Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMs

Yuanzhe Liu*, Yifan Shen*, Jingyuan Zhu, Xu Cao, Xiaofeng Zhang, Yixiao He, Wenming Ye, James Matthew Rehg, Ismini Lourentzou
NeurIPS, 2025
project page / arXiv

DeeptraceReward: Learning Human-Perceived Fakeness in Generated Videos with Multimodal LLMs

PALM: Progress-Aware Policy Learning via Affordance Reasoning for Long-Horizon Robotic Manipulation

Yuanzhe Liu, Jingyuan Zhu, Yuchen Mo, Gen Li, Xu Cao, Jin Jin, Yifan Shen, Zhengyuan Li, Tianjiao Yu, Wenzhen Yuan, Fangqiang Ding, Ismini Lourentzou
CVPR, 2026
project page / arXiv

Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMs

Yuanzhe Liu*, Yifan Shen*, Jingyuan Zhu, Xu Cao, Xiaofeng Zhang, Yixiao He, Wenming Ye, James Matthew Rehg, Ismini Lourentzou
NeurIPS, 2025
project page / arXiv

Design and source code from Jon Barron's website