Wayne Chi

Computer Science Ph.D. Student at Carnegie Mellon University
Previously, Applied Scientist at Amazon and NASA JPL

Wayne-1.jpg

waynechi[AT]cs[DOT]cmu[DOT]edu

Gates Hillman Centers 7121

Hi, my name is Wayne. I am a third year Ph.D. Student in the Computer Science Department at CMU advised by Chris Donahue and Ameet Talwalkar. I am funded by the 2025 NDSEG Fellowship.

I work on applications of machine learning that support human creativity with a focus on game development. My work includes human-centric code evaluation (1, 2) and agentic game development (3). I have also worked on music (4, 5), healthcare, and space automation (6, 7).

I completed my undergraduate and master’s at the University of Southern California. Before starting my Ph.D., I spent around 7 years in industry working as an Applied Scientist at AWS (2019-2023) and NASA’s Jet Propulsion Laboratory (2016-2019). If you are interested in any of my research topics or considering a Ph.D. after industry feel free to reach out.

news

Feb 9, 2026 EDIT-Bench is accepted to ICLR 2026 as an Oral (Top 1%)
May 1, 2025 Started my internship at Microsoft Research
May 1, 2025 Copilot Arena is accepted to ICML 2025
Apr 1, 2025 Awarded the NDSEG Fellowship
Mar 5, 2025 Preprint for Copilot Arena

selected publications

2026

  1. ICLR
    EDIT-Bench: Evaluating LLM Abilities to Perform Real-World Instructed Code Edits
    Wayne Chi, Valerie Chen, Ryan Shar, Aditya Mittal, Jenny Liang, Wei-Lin Chiang, Anastasios Nikolas Angelopoulos, Ion Stoica, Graham Neubig, Ameet Talwalkar, and Chris Donahue
    In International Conference on Learning Representations (ICLR) - Oral (Top 1%), 2026
    arXiv Website

2025

  1. ICML
    Copilot Arena: A Platform for Code LLM Evaluation in the Wild
    Wayne Chi, Valerie Chen, Anastasios Nikolas Angelopoulos, Wei-Lin Chiang, Aditya Mittal, Naman Jain, Tianjun Zhang, Ion Stoica, Chris Donahue, and Ameet Talwalkar
    In International Conference on Machine Learning (ICML), 2025
    arXiv

2021

  1. AAAI
    Symbolic music generation with transformer-gans
    Aashiq Muhamed, Liang Li, Xingjian Shi, Suri Yaddanapudi, Wayne Chi, Dylan Jackson, Rahul Suresh, Zachary C Lipton, and Alex J Smola
    In AAAI conference on artificial intelligence, 2021

2020

  1. ICAPS
    Scheduling with complex consumptive resources for a planetary rover
    Wayne Chi, Steve Chien, and Jagriti Agrawal
    In International Conference on Automated Planning and Scheduling, 2020
  2. ISMIR
    Generating Music with a Self-Correcting Non-Chronological Autoregressive Model
    Wayne Chi, Prachi Kumar, Suri Yaddanapudi, Rahul Suresh, and Umut Isik
    International Society for Music Information Retrieval, 2020
    arXiv
  3. ICML ML4MD
    Self-Correcting Non-Chronological Autoregressive Music Generation
    Wayne Chi, Prachi Kumar, Suri Yaddanapudi, Rahul Suresh, and Umut Isik
    In ICML 2020 Workshop, Machine Learning for Music Discovery (ML4MD), 2020
    arXiv

2019

  1. ICAPS
    Optimizing parameters for uncertain execution and rescheduling robustness
    Wayne Chi, Jagriti Agrawal, Steve Chien, Elyse Fosse, and Usha Guduri
    In International Conference on Automated Planning and Scheduling, 2019

2018

  1. ICAPS
    Embedding a Scheduler in Execution for a Planetary Rover
    Wayne Chi, Steve A Chien, Jagriti Agrawal, Gregg Rabideau, Edward Benowitz, Daniel M Gaines, Elyse Fosse, Stephen Kuhn, and James Biehl
    In International Conference on Automated Planning and Scheduling, 2018
  2. ICAPS
    Using Squeaky Wheel Optimization to Derive Problem Specific Control Information for a One Shot Scheduler for a Planetary Rover
    Wayne Chi, Steve Chien, and Jagriti Agrawal
    In International Conference on Automated Planning and Scheduling, 2018