Zi Li

I am a senior undergraduate in Computer Science at Nanjing University, where I work with Prof. Jingyu Hua and Prof. Yuan Zhang. From March to September 2026, I was a visiting research student at the Language Technologies Institute, Carnegie Mellon University, advised by Prof. Lei Li.

I work on AI security. With a background in applied cryptography, including work on privacy-preserving blockchain payments, my research focuses on the security and trustworthiness of large language models, especially how they can be misused or fail in realistic deployments, from secret-stealing backdoors in local fine-tuning to AI-generated peer reviews.

I am applying for Ph.D. programs starting in Fall 2027. I am always happy to discuss potential research collaborations and Ph.D. opportunities, so please feel free to reach out!

Zi Li
World Cup 2026, France 4:1 Norway, Boston
Resume GitHub
News
2026
Heading to EMNLP 2026 in Budapest to present our paper on stealing secrets from local LLM fine-tuning!
Poster Session E, Hall H, 16:45-18:15 (CET), 26 Oct.
Oct. 2026
Education
  • Nanjing University
    B.S. in Computer Science
    Aug. 2023 - present
    Nanjing, China
  • Carnegie Mellon University
    Visiting Research Student, Language Technologies Institute
    Advisor: Prof. Lei Li
    Mar. 2026 - Sep. 2026
    Pittsburgh, United States
  • University of California, Berkeley
    Visiting Student, Summer Session
    Summer 2025
    Berkeley, United States
Honors & Awards
  • National Scholarship, Ministry of Education of China
    2025
  • Zheng Gang Overseas Study Scholarship, Nanjing University
    2025
  • Best Use of Artificial Intelligence, UC Berkeley CS 160 Course Award
    2025
  • Excellence Award, Undergraduate Basic Disciplines Scholarship, Nanjing University
    2024
Publications (view details )
Secret Stealing Attacks on Local LLM Fine-Tuning through Supply-Chain Model Code Backdoors
Secret Stealing Attacks on Local LLM Fine-Tuning through Supply-Chain Model Code Backdoors

Zi Li, Tian Zhou, Wenze Li, Jingyu Hua#, Yunlong Mao#, Sheng Zhong (# corresponding author)

Proceedings of EMNLP 2026 (Main Conference)

[Paper] [Code] [Poster]

Secret Stealing Attacks on Local LLM Fine-Tuning through Supply-Chain Model Code Backdoors

Zi Li, Tian Zhou, Wenze Li, Jingyu Hua#, Yunlong Mao#, Sheng Zhong (# corresponding author)

Proceedings of EMNLP 2026 (Main Conference)

[Paper] [Code] [Poster]

PeerReveal: Learning to Generalize in AI Peer Review Detection
PeerReveal: Learning to Generalize in AI Peer Review Detection

Zi Li, André V. Duarte, Lei Li

Under review at ICLR 2027

PeerReveal: Learning to Generalize in AI Peer Review Detection

Zi Li, André V. Duarte, Lei Li

Under review at ICLR 2027

Baker: A Privacy-Preserving, NIZK-free and Efficient Payment Channel Hub Supporting Bidirectional Channels
Baker: A Privacy-Preserving, NIZK-free and Efficient Payment Channel Hub Supporting Bidirectional Channels

Wenjing Li, Zi Li, Yuan Zhang, Sheng Zhong

Proceedings of SRDS 2026

[ePrint]

Baker: A Privacy-Preserving, NIZK-free and Efficient Payment Channel Hub Supporting Bidirectional Channels

Wenjing Li, Zi Li, Yuan Zhang, Sheng Zhong

Proceedings of SRDS 2026

[ePrint]

View details
Selected Projects (view all )
Web3.0 Private Account Trading Platform
Web3.0 Private Account Trading Platform

Personal project, Dec. 2024 - Nov. 2025

[Code] [Demo]

Web3.0 Private Account Trading Platform

Personal project, Dec. 2024 - Nov. 2025

[Code] [Demo]

All projects