Chenyu Li (李晨宇)

I am a senior undergraduate student at School of Software, Tsinghua University.

My research interests primarily lie in the domains of Deep Learning and Machine Learning, with a specific focus on their applications in Time Series Analysis and Computer Vision. My long-term goal is to leverage deep learning methodologies to address real-world challenges and create impactful solutions.

Currently, I am a research intern at Courant Institute of Mathematical Sciences, New York University, advised by Prof. Saining Xie. Previously, I was in the Machine Learning Group at Tsinghua University, advised by Prof. Mingsheng Long.

Email  /  CV  /  Google Scholar  /  GitHub

• 2025-05 Our paper "PISA Experiments: Exploring Physics Post-Training for Video Diffusion Models by Watching Stuff Drop" is accepted by ICML 2025.
• 2024-10 Awarded the National Scholarship (国家奖学金).
• 2024-05 Our paper "Timer: Transformers for Time Series at Scale" is accepted by ICML 2024.
• 2023-09 Our paper "Koopa: Learning Non-stationary Time Series Dynamics with Koopman Predictors" is accepted by NeurIPS 2023.

• Bachelor in Software Engineering, 2021-2025
  School of Software, Tsinghua University
  GPA: 3.89/4.0

• Apache IoTDB
  Develop IoTDB Artificial Intelligence Node, a native machine learning engine of time series database. Users build, train, manage and use machine learning models in IoTDB databases using SQL statements.

• KDD Excellent Reviewer (Top 10%), KDD Program Chairs, 2025
• Outstanding Graduates of Beijing (Top 1%), Beijing Municipal Education Commission, 2025
• Outstanding Graduates of Tsinghua (Top 1%), Tsinghua University, 2025
• National Scholarship (国家奖学金, top scholarship in China, 0.2% domestically), Ministry of Education, 2024
• SenseTime Scholarship (30 undergraduates domestically), SenseTime, 2023
• Huawei Scholarship (华为奖学金), Tsinghua University, 2023
• 12•9 Scholarship (一二•九奖学金, top 1%), Tsinghua University, 2022

• Reviewer: NeurIPS (2025), ICML (2025), KDD (2025)