Eric Lei

New York, NY

I am a research scientist at JPMorganChase, Global Technology Applied Research, in NYC. I completed my Ph.D. at the University of Pennsylvania in the Electrical and Systems Engineering department, where I was advised by Prof. Shirin Saeedi Bidokhti and Prof. Hamed Hassani, and supported by a NSF Graduate Research Fellowship. Prior to that, I obtained my B.S. from Cornell University.

Education

University of Pennsylvania
Ph.D. in Electrical and Systems Engineering, 2025

Cornell University
B.S. in Electrical and Computer Engineering, 2020

selected publications

  1. Preprint
    Optimal Neural Compressors for the Rate-Distortion-Perception Tradeoff
    Eric Lei, Hamed Hassani, and Shirin Saeedi Bidokhti
    arXiv preprint arXiv:2503.17558 2025
    HTML
  2. ICLR
    Approaching Rate-Distortion Limits in Neural Compression with Lattice Transform Coding
    Eric Lei, Hamed Hassani, and Shirin Saeedi Bidokhti
    International Conference on Learning Representations 2025
    Spotlight
    HTML Code
  3. ICLR
    PaLD: Detection of Text Partially Written by Large Language Models
    Eric Lei, Hsiang Hsu, and Chun-Fu Chen
    International Conference on Learning Representations 2025
    HTML Code
  4. ICIP
    WrappingNet: Mesh Autoencoder via Deep Sphere Deformation
    Eric Lei, Muhammad Asad Lodhi, Jiahao Pang, Junghyun Ahn, and Dong Tian
    IEEE International Conference on Image Processing (ICIP) 2024
    Best Student Paper Award
    HTML Code
  5. ICML NCW
    Text + Sketch: Image Compression at Ultra Low Rates
    Eric Lei, Yiğit Berkay Uslu, Hamed Hassani, and Shirin Saeedi Bidokhti
    ICML Workshop on Neural Compression 2023
    HTML Code
  6. ICLR
    On a Relation Between Rate-Distortion Theory and Optimal Transport
    Eric Lei, Hamed Hassani, and Shirin Saeedi Bidokhti
    International Conference on Learning Representations (Tiny Papers Track) 2023
    HTML
  7. JSAIT
    Neural Estimation of the Rate-Distortion Function With Applications to Operational Source Coding
    Eric Lei, Hamed Hassani, and Shirin Saeedi Bidokhti
    IEEE Journal on Selected Areas in Information Theory 2023
    HTML Code
  8. L4DC
    Robust Graph Neural Networks via Probabilistic Lipschitz Constraints
    Raghu Arghal, Eric Lei, and Shirin Saeedi Bidokhti
    In Proceedings of The 4th Annual Learning for Dynamics and Control Conference 2022
    HTML
  9. ICML
    Out-of-Distribution Robustness in Deep Learning Compression
    Eric Lei, Hamed Hassani, and Shirin Saeedi Bidokhti
    ICML Workshop on Information-Theoretic Methods for Rigorous, Responsible, and Reliable Machine Learning 2021
    selected for 1 of 4 contributed talks
    HTML