Kevin Zhou

prof_pic.jpg

Hi! I am currently a Machine Learning Scientist at Abridge, working on AI to help improve healthcare and the lives of clinicians.

Previously, I was an MSc Advanced Computing student at Imperial College London where I had the pleasure of being supervised by Professor Marek Rei and Dr. Kris Cao of Cohere for my dissertation on efficient optimal data mixing strategies for pre-training language models. I also worked with Professor Francesca Toni to research large language model uncertainty quantification methods and their integration with the argumentative large language model framework. Before Imperial, I studied Computer Science and Mathematics at Cornell University where I was fortunate to be able to work with Professors Wen Sun and Nathan Kallus on distributional reinforcement learning.

Publications

2025

  1. EMNLP Findings
    Evaluating Uncertainty Quantification Methods in Argumentative Large Language Models
    Kevin Zhou, Adam Dejl, Gabriel Freedman, Lihu Chen, Antonio Rago, and Francesca Toni
    In Findings of the Association for Computational Linguistics: EMNLP 2025, Nov 2025

2023

  1. NeurIPS
    The Benefits of Being Distributional: Small-Loss Bounds for Reinforcement Learning
    Kaiwen Wang, Kevin Zhou, Runzhe Wu, Nathan Kallus, and Wen Sun
    Advances in Neural Information Processing Systems 37, Dec 2023