Yaoxuan Wu

ai-fused.PNG

I am a fifth-year PhD candidate in Computer Science at UCLA, advised by Prof. Miryung Kim. I obtained my B.Sc., Summa Cum Laude in Computer Science from Turing Class, Peking University.

Previously, my research focused on improving test naturalness, API modeling, and test oracles for symbolic execution and fuzzing, with a particular emphasis on data-intensive computing systems. More recently, I have shifted my focus to research on AI agents and agentic systems. I am always happy to explore potential collaborations—please feel free to reach out!

Selected publications

* Equal contribution.

  1. ICSE 2027
    Agentic Proof and Property-Based Testing via Property-Templates in Data-Intensive Computing
    Seongmin Lee*, Yaoxuan Wu*, and Miryung Kim
    In IEEE/ACM International Conference on Software Engineering (ICSE), to appear, 2027
    arXiv
  2. Preprint
    Properties with Holes: Meta-Property-Based Testing of Data-Intensive Computing Frameworks
    Yaoxuan Wu, Ingrid Lee, Ahmad Humayun, Muhammad Ali Gulzar, and Miryung Kim
    arXiv preprint arXiv:2606.11132, 2026
    arXiv
  3. ICPC 2026
    PALM: Path-aware LLM-based Test Generation with Comprehension
    Yaoxuan Wu, Xiaojie Zhou, Ahmad Humayun, Muhammad Ali Gulzar, and Miryung Kim
    The 34th IEEE/ACM International Conference on Program Comprehension, 2026
    arXiv PDF Video
  4. FSE 2024
    Natural Symbolic Execution-based Testing for Big Data Analytics
    Yaoxuan Wu, Ahmad Humayun, Muhammad Ali Gulzar, and Miryung Kim
    Proceedings of the ACM on Software Engineering, 2024
    PDF