Hello👋! I am Mingzhe Li(李明哲), a first-year Ph.D student at University of Massachusetts, Amherst, advised by Prof. Shiqing Ma.

I received my B.Eng in Cyberspace Security from Wuhan University in 2025. Before my time at WHU, I completed three years of study at the No.1 Middle School Affiliated to Central China Normal University （华中师大一附中）.

My research focuses on artificial intelligence security and privacy, Responsible Generative AI, and Agentic AI.

You can find my CV here (updated May 20, 2026).

🔥 News

• 2026.08   🎉🎉 1 paper is accepted to EMNLP 2026.
• 2026.07   🎉🎉 1 paper is accepted to COLM 2026.
• 2026.05  🚀🚀 CiteTracer released — code and dataset are now open-sourced.
• 2026.02   🎉🎉 PROMPTMINER is accepted to CVPR 2026. See you in Denver!
• 2025.11  🚀🚀 PROMPTMINER released — code is now open-sourced.
• 2025.04  🏆🏆 I am selected as an outstanding undergraduate graduate.
• 2025.01:  🎉🎉 PAPILLON is accepted to Usenix Security 2025.
• 2024.12:  🎉🎉 1 paper is accepted to IEEE TIFS 2025.

📝 Selected Publications

For the full publication list, please see my CV.

Preprint

Source or It Didn’t Happen: A Multi-Agent Framework for Citation Hallucination Detection

Mingzhe Li, Zhiqiang Lin, Shiqing Ma

Preprint

Code

Preprint

Membership Inference Attacks on Vision-Language-Action Models

Yuefeng Peng*, Mingzhe Li*, Kejing Xia, Renhao Zhang, Amir Houmansadr

Preprint, * indicates equal contribution

EMNLP 2026

Disappearing Ink: Obfuscation Breaks N-gram Code Watermarks in Theory and Practice

Gehao Zhang, Mingzhe Li, Eugene Bagdasarian, Shiqing Ma, Juan Zhai

Conference on Empirical Methods in Natural Language Processing (EMNLP 2026, Findings)

Code

CVPR 2026

PROMPTMINER: Black-Box Prompt Stealing against Text-to-Image Generative Models via Reinforcement Learning and Fuzz Optimization

Mingzhe Li, Renhao Zhang, Zhiyang Wen, Siqi Pan, Bruno Castro da Silva, Juan Zhai, Shiqing Ma

Conference on Computer Vision and Pattern Recognition (CVPR 2026, main)

Code

Preprint

EDITOR: Effective and Interpretable Prompt Inversion for Text-to-Image Diffusion Models

Mingzhe Li, Kejing Xia, Gehao Zhang, Zhenting Wang, Guanhong Tao, Siqi Pan, Juan Zhai, Shiqing Ma

Preprint

Code

USENIX Security 2025

PAPILLON: Efficient and Stealthy Fuzz Testing-Powered Jailbreaks for LLMs

Xueluan Gong*, Mingzhe Li*, Yilin Zhang, Fengyuan Ran, Chen Chen, Yanjiao Chen, Qian Wang, Kwok-Yan Lam

34th Usenix Security Symposium (USENIX Security 2025), * indicates equal contribution

Code

TIFS 2025

Augmenting Model Extraction Attacks against Disruption-Based Defenses

Xueluan Gong, Shuaike Li, Yanjiao Chen, Mingzhe Li, Rubin Wei, Qian Wang and Kwok Yan Lam

IEEE Transactions on Information Forensics & Security (TIFS 2025)

🎖 Honors and Awards

• Outstanding Undergraduate Graduate, WHU Apr. 2025
• National Second Prize in Contemporary Undergraduate Math Contest in Modeling, Top 2% nationwide Nov. 2023
• Second Prize in the Chinese Collegiate Computing Competition Jun. 2023
• Second Prize in the “Huazhong Cup” College Student Mathematics Modeling Challenge May 2023
• Third Prize in the Hubei Province College Student Mathematics Competition Mar. 2023
• “Merit Student” Honorary Title, Twice Sept. 2023 & Sept. 2022
• “Active Participant in Social Activities” Honorary Title Jun. 2022

📖 Educations

• 2025.09 - Present, Ph.D Student, University of Massachusetts, Amherst
• 2021.09 - 2025.06, Undergraduate, School of Cyber Science and Engineering, Wuhan University
• 2018.09 - 2021.06, Senior Middle School Student, the No.1 Middle School Affiliated to Central China Normal University

💻 Internships

• 2025.02 - 2025.06, ByteDance, Shenzhen, China.