Xiaoyu Xiong | Tsinghua University

About Me

I am an undergraduate student in Computer Science at the Institute for Interdisciplinary Information Sciences (IIIS, Yao Class), Tsinghua University, majoring in Computer Science. Prior to this, I was a member of the Chinese national team for the Asian Physics Olympiad.

In terms of academic performance, my average GPA in major courses is 3.93. I began my research under the supervision of Prof. Tao Du, focusing on physical AI and robotics. During the Spring and Summer of 2026, I was a student intern at Carnegie Mellon University, advised by Prof. Minchen Li, where I worked on generative simulation research.

My primary research goal is to leverage physical knowledge to build more usable and robust AI systems capable of performing realistic and practical tasks.

Research Interests

I am broadly interested in Physical AI: building AI systems that understand, generate, and act in the physical world through simulation and learning. My work approaches this goal by automatical environment construction, precise information sensing, and optimized task actions, building a complete framework to enable AI agents to interact with the real world.

News

Apr. 2026 Learning to Reconfigure was accepted to ICML 2026.
Jun. 2025 Awarded the SenseTime Scholarship, with only 30 recipients nationwide per year.
Jan. 2025 TopoGaussian was accepted to ICLR 2025.