Yu He

swan.PNG

Hi, I am Heyu, a third-year CS PhD student at Stanford, where I am advised by Ellen Vitercik. Previously, I graduated from University of Cambridge with BA+MEng in Computer Science, where I was supervised by Pietro Liò.

I am interested in understanding, evaluating, and improving the algorithmic and structured reasoning abilities in neural networks, such as large language models and graph neural networks.

My research is generously supported by the Cubist PhD Fellowship and previously a Stanford School of Engineering Fellowship.

research

  1. nar-position-paper.png
    Neural Algorithmic Reasoning Must Explain When Neuralization Adds Value
    Yu He, Robert R Nerem, Timo Stoll, Semih Cantürk, Dobrik Georgiev Georgiev, Solveig Wittig, Chendi Qian, Floris Geerts, Stefanie Jegelka, Ellen Vitercik, Yusu Wang, Nikolaos Karalias, Christopher Morris
    In 3rd AI for Math Workshop: Toward Self-Evolving Scientific Agents, 2026
    Paper
  2. dsr.png
    Can LLMs Reason Structurally? Benchmarking via the Lens of Data Structures
    Yu He*, Yingxi Li*, Colin White, Ellen Vitercik
    In International Conference on Machine Learning (ICML), 2026
    Paper Code
  3. primal-dual.png
    Primal-Dual Neural Algorithmic Reasoning
    Yu He and Ellen Vitercik
    In International Conference on Machine Learning (ICML), 2025
    Spotlight (top 2.6%)
    Paper Code Poster
  4. rewire.png
    Overcoming Information Bottlenecks in Directed Graph Neural Networks through Rewiring
    Yu He, Ishani Karmarkar, and Ellen Vitercik
    In Learning on Graph Conference (LoG), 2025
    Paper
  5. dem.png
    Deep Equilibrium Models For Algorithmic Reasoning
    Sophie Xhonneux, Yu He, Andreea Deac, Jian Tang, Gauthier Gidel
    In The Third Blogpost Track at International Conference on Learning Representations (ICLR), 2024
    Paper Blog
  6. higher-egp.png
    Higher-Order Expander Graph Propagation
    Thomas Christie* and Yu He*
    In NeurIPS 2023 Workshop: New Frontiers in Graph Learning, 2023
    Paper Code Poster
  7. sheaf-pe.png
    Sheaf-based Positional Encodings for Graph Neural Networks
    Yu He, Cristian Bodnar, and Pietro Liò
    In NeurIPS 2023 Workshop on Symmetry and Geometry in Neural Representations, 2023
    Paper Poster
  8. cnap.png
    Continuous Neural Algorithmic Planners
    Yu He, Petar Veličković, Pietro Liò, Andreea Deac
    In Proceedings of the First Learning on Graphs Conference, 2022
    Paper Code Poster