Shaoxuan Li 李少轩

About

I am a third-year undergraduate studying Computer Science and Technology at Tsinghua University, with a research focus on embodied intelligence.

I am particularly interested in learning effective visual representations for embodied systems and making embodied models more computationally efficient. I'm also interested in AI agents(because we are using it everyday!) and reinforcement learning.

Feel free to reach out at lsx24@mails.tsinghua.edu.cn.

News

• Jun 18th 2026 PerceptionComp is accepted to ECCV 2026!
• Mar 27th 2026 PerceptionComp is released on arXiv.
• Oct 1st 2025 Awarded the National Scholarship (top 5% of Tsinghua students).

Publications

^* denotes equal contribution.

PerceptionComp: A Video Benchmark for Complex Perception-Centric Reasoning

Shaoxuan Li*, Zhixuan Zhao*, Hanze Deng*, Zirun Ma*, Shulin Tian, Zuyan Liu, Yushi Hu, Haoning Wu, Yuhao Dong, Benlin Liu, Ziwei Liu, Ranjay Krishna

ECCV, 2026

Project / arXiv / Code

Research Experience

• Jul – Sep 2026 Intern, Motubrain Team · Worked on embodied foundation models, focusing on model architecture, WAM distillation, and inference acceleration.
• Jun – Aug 2026 Intern, TSAIL · Studied reinforcement learning for diffusion-based planners in autonomous driving.
• Feb – Jun 2026 Intern, THUNLP · Developed AutoSFT, an agentic system for automating supervised fine-tuning (SFT).

Education

• 2024 – Present Tsinghua University · B.E. in Computer Science and Technology · GPA 3.96/4.00, top 2%

Honors & Awards

• 2025 National Scholarship — top 5% of Tsinghua students