About Me

I am the fourth-year Ph.D. Candidate from School of Information @ University of Michigan-Ann Arbor, supervised by Dr. V.G.Vinod Vydiswaran.

Research

My research interests are Cyber Security (red-teaming, backdoor attack & defense of LLMs), Generative models, Health informatics.

Education

• Ph.D. in Informatics, University of Michigan , 2020 - 2024 July (expected)
• M.S. in Electrical Computer Engineering, University of Michigan,2017 - 2019
• B.S. in Electrical Engineering, Nankai University, China,2013 - 2017

Industry Experience

• 05/2023 - 08/2023 : Research Scientiest Intern @ Yahoo Research
• 10/2019 - 08/2020 : Research Associate @ Michigan Medicine

Teaching Experience

Graduate Student Instructor:

• WN 2024: SI 630 Natura Language Processing Algorithms and People.
  
• WN 2023: LHS 712 Natural Language Processing for Health.
  

Public Service

• Reviewer or PC of conference: ACL ARR, EACL 2023, EMNLP 20’21’22’23
  
• Reviewer of Journal: Frontiers in Big Data, section Cybersecurity and Privacy.
  

Awards

• Rackham Conference Travel Grant
  
• UMSI Conference Travel Grant
  

Selected Publications

• [Cyber Security] ChatGPT as an Attack Tool: Stealthy Textual Backdoor Attack via Blackbox Generative Model Trigger. [Paper]
  Jiazhao Li, Yijin Yang, Zhuofeng Wu, V.G. Vinod Vydiswaran, Chaowei Xiao
  NAACL 2024.
  

• [Cyber Security] Defending against Insertion-based Textual Backdoor Attacks via Attribution. [Paper]
  Jiazhao Li, Zhuofeng Wu, Wei Ping, Chaowei Xiao, V.G.Vinod Vydiswaran
  Findings of ACL 2023.
  

• [Machine Translation] PharmMT: A Neural Machine Translation Approach to Simplify Prescription Directions. [Paper] [BibTex][Video]
  Jiazhao Li, Corey Lester, Xinyan Zhao, Yuting Ding, Yun Jiang, and V.G.Vinod Vydiswaran.
  Findings of EMNLP 2020.
  

• [Information Retrieval] Re-ranking biomedical literature for precision medicine with pre-trained neural models. [Paper] [BibTex]
  Jiazhao Li, Adharsh Murali, Qiaozhu Mei, V.G.Vinod Vydiswaran.
  ICHI 2020.