YZ

Yuan Zhuang 庄元

I am a Ph.D. student in Computer Science and Engineering at the University of Connecticut. My primary advisor is Prof. Fei Miao, and my co-advisors are Prof. Dongjin Song and Prof. Jonathan Shihao Ji.

My research focuses on efficient foundation models and adaptive computation, including long-context language modeling and mixture-of-experts. I also work on reinforcement learning and embodied AI.

In summer 2026, I interned at Microsoft Research, mentored by Dr. Xiaodong Liu, developing adaptive-depth methods for test-time-training language models. Previously, I worked on vision-language-action models at JD.com. I received my M.S. in Data Science from the University of Pennsylvania and my B.Eng. in Ocean Engineering and Technology from Zhejiang University.

I expect to graduate in December 2026 and am seeking full-time Research Scientist, Research Engineer, and Applied Scientist opportunities.

GitHub Google Scholar LinkedIn CV
Yuan Zhuang standing by a lake with mountains in the background

Selected Research

2026 NeurIPS

Low-Rank Adaptation for Critic Learning in Off-Policy Reinforcement Learning

Yuan Zhuang*, Yuexin Bian*, Sihong He, Jie Feng, Qing Su, Songyang Han, Jonathan Petit, Shihao Ji, Yuanyuan Shi, Fei Miao

arXiv Code
2026 ICLR

LD-MoLE: Learnable Dynamic Routing for Mixture of LoRA Experts

Yuan Zhuang*, Yi Shen*, Yuexin Bian, Qing Su, Shihao Ji, Yuanyuan Shi, Fei Miao

arXiv Code
2025 IROS

YOLO-MARL: You Only LLM Once for Multi-Agent Reinforcement Learning

Yuan Zhuang*, Yi Shen*, Zhili Zhang, Yuxiao Chen, Fei Miao

arXiv Code
2026 CoRL

Artificial Foveated Perception for Mitigating Shortcut Learning in Robotic Foundation Models

Xiatao Sun, Yuan Zhuang, Mateo Sanchez Lopez Negrete, Matei-Victor Coldea, Chen Liang, Haoyang Zhang, Che Liu, Ziyao Zeng, Shawn Li, Qian Wang, Fei Miao, Daniel Rakita

arXiv Code

Work Experience

- Redmond, WA

Microsoft Research

Research Intern

Developed adaptive-depth methods for long-context test-time-training language models.

- Beijing, China

JD.com

Research Intern, Embodied Technology Group

Studied generalization and robustness of vision-language-action models on public benchmarks within a logistics research project.

Contact

paulyuanzhuang@gmail.com