About me

I will join Department of Computer Science and Engineering at University of California, Riverside (CSE@UCR) as an Assistant Professor. I am looking for motivated students to join my group (including PhD students to start in 2025~2026).

I am currently a final-year Ph.D. Candidate at UCLA Computer Science Department, advised by Prof. Cho-Jui Hsieh. Prior to UCLA, I received my B.Eng. degree from the CST department at Tsinghua University.

My primary research focus is on trustworthy machine learning, and I am broadly interested in developing more trustworthy and reliable AI models. In particular, I work on verifiable machine learning, with topics such as formal verification for neural networks, training verification-friendly neural networks with stronger verifiable guarantees, applications of verifiable machine learning in mission-critical scenarios, and more recently, verification for generative AI. Additionally, I also study empirical methods for evaluating and improving the robustness or safety of large-scale ML foundation models.

Selected Publications

(* Equal contribution; for full publications: see Publications or Google Scholar )
Neural Network Verification with Branch-and-Bound for General Nonlinearities
Lyapunov-stable Neural Control for State and Output Feedback: A Novel Formulation
Defending LLMs against Jailbreaking Attacks via Backtranslation
Red Teaming Language Model Detectors with Language Models
Effective Robustness against Natural Distribution Shifts for Models with Different Training Data
Fast Certified Robust Training with Short Warmup
Automatic Perturbation Analysis for Scalable Certified Robustness and Beyond

Software

  • auto_LiRPA : A library for automatically computing differentiable verified output bounds for general computational graphs. Originally proposed in our NeurIPS 2020 paper, it is the core bound computation engine in the award-winning alpha-beta-CROWN.
  • alpha-beta-CROWN : A comprehensive neural network verification toolbox that consists of multiple complete verification algorithms (such as GenBaB) on top of auto_LiRPA. It is the 1st place winner at the annual VNN-COMP from 2021 to 2024.

Awards

  • UCLA Dissertation Year Award (fellowship), 2024-2025
  • Amazon Fellowship (Amazon & UCLA Science Hub fellowship), 2022-2023
  • 4X first-place winner at the International Verification of Neural Networks Competition (VNN-COMP), 2021-2024

Teaching

TA at UCLA: