Research Profile

Research interests: Human-AI interaction; human-centered AI evaluation and auditing; AI privacy, security, and governance; LLM and computer-use agents; human oversight and information practices.

I study how people understand, interact with, and govern increasingly autonomous AI systems. My research develops empirical methods and interactive systems for evaluating AI agents across privacy, security, reliability, and human-oversight risks, with particular attention to how information is disclosed, interpreted, and acted upon in human-AI interactions. By combining large-scale behavioral analysis, controlled experiments, benchmark development, and human-centered system design, I seek to build sociotechnical foundations for AI systems that are understandable, auditable, and governable.

Academic Appointments

09/2026 -

Postdoctoral Fellow

Johns Hopkins University, Baltimore, MD

Whiting School of Engineering.

Advisor: Yaxing Yao

Education

2022 - 2026

Ph.D. in Computer Science and Engineering

University of Notre Dame, Notre Dame, IN

Department of Computer Science and Engineering.

Advisors: Toby Jia-Jun Li and Yanfang Ye

2021 - 2022

M.S. in Educational Technology and Applied Learning Sciences

Carnegie Mellon University, Pittsburgh, PA

Human-Computer Interaction Institute, School of Computer Science.

2015 - 2020

B.Eng. in Industrial Design

Tongji University, Shanghai, China

College of Design and Innovation.

Experience

05/2026 - 07/2026

Applied Scientist Intern

Amazon, Seattle, WA

Led the development of Bench2Robust, a framework for evaluating tool-using AI agents under transient, persistent, and silent failures across more than 1,000 tasks. Identified a substantial gap between clean-task performance and failure-aware recovery, and developed Bayesian tool-reliability modeling and reinforcement-learning strategies to improve failure recovery while preserving performance when tools operate normally.

Hosts: Vy Nguyen and Ziji Zhang

07/2025 - 04/2026

Student Researcher

Google, Sunnyvale, CA

Led the development of behavioral canaries, a method for auditing whether private retrieved context influences reinforcement-learning fine-tuning without relying on verbatim memorization. Designed controlled evaluations across retrieval-based tasks to measure privacy auditing performance while preserving downstream model utility; the resulting paper was accepted to COLM 2026.

Hosts: Dayu Yuan and Peter Kairouz

10/2024 - 12/2024

Research Intern

Max Planck Institute for Security and Privacy, Bochum, Germany

Research focus: literature synthesis and research agenda development on privacy risks, human oversight, and security challenges in LLM-powered GUI agents.

Host: Abraham Mhaidli

Selected Fellowships, Programs, Honors, and Awards

  • Aspen Cyber Summit Scholar, 2026
  • International Programme on AI Evaluation: Capabilities & Safety, Inaugural Cohort, 2026
  • Advanced Researcher Award, University of Notre Dame, 2026
  • Travel Grant, Symposium on Usable Privacy and Security (SOUPS), 2025
  • Third prize of Tech Ethics Hackathon ($1,000), Notre Dame-IBM Technology Ethics Lab, 2024
  • Gary Marsden Travel Award, ACM SIGCHI, 2022
  • Best Implementation Award ($2,000), MobileHCI 2022
  • Merit Scholarship ($7,000), Carnegie Mellon University

Publications

Preprints

Journal Papers

Conference Papers(* indicates equal contribution.)

[C.9]
2026

NARRA-Gym for Evaluating Interactive Narrative Agents

Yue Huang, Yuchen Ma, Jiayi Ye, Wenjie Wang, Zipeng Ling, Xingjian Hu, Yuexing Hao, Zichen Chen, Zhangchen Xu, Yunhong He, Zhengqing Yuan, Yujun Zhou, Kehan Guo, Chaoran Chen, Toby Jia-Jun Li, Stefan Feuerriegel, and Xiangliang Zhang

The Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS 2026), Evaluations and Datasets Track

Workshop Papers

[W.6]
2026

Beyond Task Completion: A Process-Oriented Framework for Evaluating Preference and Value Alignments in Computer-Using Agents

Yukun Yang, Simret Araya Gebreegziabher, Hojun Yoo, Charles Chiang, Chaoran Chen, Annalisa Szymanski, Hyo Jin Do, Zahra Ashktorab, Werner Geyer, Diego Gómez-Zará, and Toby Jia-Jun Li

IUI 2026 Workshop on Trustworthy Computer-Using Generalist Agents for Intelligent User Interfaces

Lightly-Reviewed Publications and Extended Abstracts

Teaching Experience

Spring 2025

Teaching Assistant, CSE 40567: Computer Security

Department of Computer Science and Engineering, University of Notre Dame

Fall 2023

Teaching Assistant, CSE 40625: Machine Learning

Department of Computer Science and Engineering, University of Notre Dame

Fall 2022

Teaching Assistant, 05-738: Evidence-Based Educational Design

Human-Computer Interaction Institute, Carnegie Mellon University

Spring 2022

Teaching Assistant, 05-440/05-640: Interaction Techniques

Human-Computer Interaction Institute, Carnegie Mellon University

Student Mentoring

Undergraduate Students at Notre Dame

  • Alfonso Alcaraz Moreno, B.S. in CS, 2026
  • Louis Cao, B.S. in CS, 2025
  • Leyang Li, B.S. in CS, 2024; first position Ph.D. student at JHU
  • Luke Cao, B.S. in CS, 2024; first position SDE at Google

Students outside Notre Dame

  • Zixin Wang, Ph.D. in CS, UMich, 2026
  • Jiayi Ye, B.S. in CS, Sichuan University, 2025
  • Bingcan Guo, M.S. in HCDE, U. Washington, 2025; first position Ph.D. student at UCLA
  • Daodao Zhou, M.S. in CS, Virginia Tech, 2024

Visiting Students

  • Xujia Liu, U. Washington, Summer 2026
  • Jiawen Liu, Fudan University, Summer 2026
  • Jingyu Tang, Huazhong University of Science and Technology, Summer 2025; first position Ph.D. student at UT Austin
  • Yan Xu, Nanjing University, Summer 2024; first position M.S. student at UCSD
  • Tianlu Zhou, Tsinghua University, Summer 2024; first position M.S. student at U. Washington
  • Wenxin Song, Chinese University of Hong Kong, Summer 2023; first position M.S. student at Columbia University
  • Weijun Li, Zhejiang University, Summer 2023; first position Ph.D. student at UC Irvine

Academic Service

Organizer

  • Human-Agent Collaboration Workshop at CHI 2026

Associate Chair / Area Chair

  • CHI 2027
  • IAB Workshop at NeurIPS 2026

Conference Reviewer

  • CHI 2023-2026
  • CSCW 2025
  • IUI 2025-2026
  • UIST 2026
  • Special recognitions for outstanding reviews at CHI 2024 LBW, CSCW 2025, IUI 2025-2026, CHI 2026, and UIST 2026

Workshop Reviewer

  • HAIPS Workshop at CCS 2025
  • Trustworthy AI Workshop at ICLR 2026
  • HAIPS Workshop at COLM 2026

Artifact Evaluation Committee

  • IEEE S&P 2027

Journal Reviewer

  • ACM TOPS
  • ACM TAISAP
  • ACM AI Letters
  • IJHCS
  • PACM EICS