Yingxi Li

prof_pic.jpg
yingxi at stanford dot edu

Hi, I’m Yingxi Li, a forth year PhD candidate in Operations Research, Management Science and Engineering (MS&E) department at Stanford University. I am fortunate to be advised by Professor Ellen Vitercik. Previously, I graduated with a B.S. from the Operations Research and Information Engineering (ORIE) department of Cornell University, where I was also lucky enough to work with Professor Madeleine Udell and Professor Shane Henderson.

My interest sits at the intersection of artificial intelligent, algorithm, and discrete optimization. I am particularly interested in using machine learning to design faster, more scalable, and robust methods for discrete optimization problems. I am also interested in algorithmic reasoning capabilities of large language models (LLMs) with an eye toward leveraging them for algorithm design.

My research is generously supported by the Amazon AI Fellowship and Stanford MS&E departmental fellowship.

Publication and Preprints

* means equal contribution. For theory papers, authors are always listed alphabetically by last name.

  1. Can LLMs Reason Structurally? An Evaluation via the Lens of Data Structures
    Yu He*, Yingxi Li*, Colin White, and Ellen Vitercik
    In International Conference on Machine Learning (ICML), 2026
    arXiv Dataset Code
  2. Smoothed Analysis of Online Metric Matching with a Single Sample: Beyond Metric Distortion
    Yingxi Li, Ellen Vitercik, and Mingwei Yang
    In Innovations in Theoretical Computer Science (ITCS), 2026
    arXiv
  3. Compositional Reasoning in Language Models under Reinforcement Learning Post-Training
    Yu He, Yingxi Li, Yifei Wang, and Ellen Vitercik
    In Conference on Neural Information Processing Systems (NeurIPS), 2026
    arXiv
  4. Accelerating data-driven algorithm selection for combinatorial partitioning problems
    Vaggos Chatziafratis, Ishani Karmarkar, Yingxi Li, and Ellen Vitercik
    In Conference on Neural Information Processing Systems (NeurIPS), spotlight (top 3% of all submissions), 2025
    arXiv Code
  5. LLMs for Cold-Start Cutting Plane Separator Configuration
    Connor Lawless*, Yingxi Li*, Anders Wikum*, Madeleine Udell, and Ellen Vitercik
    In Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research (CPAIOR), 2025
    arXiv Code

Miscellaneous

You can pronounce my name EENG-shee or yìng-xī (映溪). I went by Diana in high school and during my first year of college, and it remains my Starbucks name.

Outside of research, I love music and rock climbing! In an alternate life where I know nothing about optimization or machine learning, I'd be a street accordionist, playing cheerful folk tunes for people passing by. (I don't actually know how to play the accordion.)