Xutong Zhao

Xutong Zhao

PhD student at Mila - Quebec AI Institute & Polytechnique Montréal

About me

I am a PhD student at Mila - Quebec AI Institute and Polytechnique Montréal, advised by Prof. Sarath Chandar. I completed my M.Sc. at the University of Alberta, advised by Prof. Adam White. Previously I interned at Meta AI advised by Chen Zhu, Huawei Noah’s Ark Lab advised by Yangchen Pan, and Google Deepmind advised by Daniel Toyama.

My research interests span reinforcement learning, multi-agent systems, and foundation models. I am particularly interested in developing autonomous learning agents that interact with their environments and other agents to accomplish complex tasks.

Publications

  • Learning Awareness for Effective and Efficient Learning in Social Dilemmas
    Xutong Zhao, Ziyan Luo, Tianwei Ni, Yaqi Xie, Wanru Zhao, Xujie Si, Doina Precup, Sarath Chandar
    In submission to International Conference on Learning Representations (ICLR) 2027. Manuscript available upon request.

  • Boosting LLM Reasoning via Spontaneous Self-Correction
    Xutong Zhao, Tengyu Xu, Xuewei Wang, Zhengxing Chen, Di Jin, Liang Tan, Yen-Ting Lin, Zishun Yu, Zhuokai Zhao, Yun He, Sinong Wang, Han Fang, Sarath Chandar, Chen Zhu
    Conference on Language Modeling (COLM). October 2025. paper
    AI for Math Workshop (AI4MATH) @ International Conference on Machine Learning (ICML). July 2025.

  • Multi-level Advantage Credit Assignment for Cooperative Multi-Agent Reinforcement Learning
    Xutong Zhao, Yaqi Xie
    International Conference on Artificial Intelligence and Statistics (AISTATS). May 2025.
    paper, code

  • Conditionally Optimistic Exploration for Cooperative Deep Multi-Agent Reinforcement Learning
    Xutong Zhao, Yangchen Pan, Chenjun Xiao, Sarath Chandar, Janarthanan Rajendran
    Conference on Uncertainty in Artificial Intelligence (UAI). August 2023.
    paper, code

  • Towards Few-shot Coordination: Revisiting Ad-hoc Teamplay Challenge In the Game of Hanabi
    Hadi Nekoei, Xutong Zhao, Janarthanan Rajendran, Miao Liu, Sarath Chandar
    Conference on Lifelong Learning Agents (CoLLAs). August 2023.
    paper, code

  • No More Pesky Hyperparameters: Offline Hyperparameter Tuning for RL
    Han Wang, Archit Sakhadeo, Adam White, James Bell, Vincent Liu, Xutong Zhao, Puer Liu, Tadashi Kozuno, Alona Fyshe, Martha White
    Transactions on Machine Learning Research (TMLR). August 2022.
    paper

  • An Empirical Study of Model-Free Exploration for Deep Reinforcement Learning
    Master of Science thesis. November 2021.
    link

Experience

  • Research Intern @ Meta AI, Oct.2024 - Feb.2025
  • Research Assistant @ Mila ‐ Quebec AI Institute & Polytechnique Montréal, Sep.2021 - present
  • Research Intern @ Huawei Noah’s Ark Lab, Aug.2022 - Dec.2022
  • Research Assistant @ University of Alberta, Sep.2019 - Aug.2021
  • Research Intern @ Google DeepMind, Jun.2019 - Aug.2019

Teaching

  • IFT 6135B ‐ Representation Learning, Université de Montréal. Winter 2026
  • INF8953DE - Reinforcement learning, Polytechnique Montréal. Fall 2021
  • CMPUT 397 - Reinforcement learning, University of Alberta. Winter 2020, Fall 2019