Zongmin Zhang
HKUST (Guangzhou) · Guangzhou, China · zzhang498@connect.hkust-gz.edu.cn
Hello! I am a first-year Ph.D. student in Data Science and Analytics at The Hong Kong University of Science and Technology (Guangzhou), advised by Prof. Xinlei He and Prof. Jiaheng Wei.
My research interests are in AI security & privacy, with a focus on vision segmentation models, LLMs/VLLMs, and image watermarking. In particular, I study backdoor attacks and jailbreaking of modern foundation models, as well as attack/defense for image watermarking systems.
Before that, I received my M.Sc. in Computer Science from City University of Hong Kong and my B.Sc. in Computer Science (Database Systems) from The University of New South Wales.
News
| 2025.11 | |
|---|---|
| 2025.09 | |
| 2025.05 | |
Selected Publications
* indicates equal contribution.
- arXiv’25Backdoor Attacks on Prompt-Driven Video Segmentation Foundation ModelsarXiv preprint Dec 2025
- arXiv’25To Survive, I Must Defect: Jailbreaking LLMs via the Game-Theory ScenariosarXiv preprint Nov 2025
- AAAI’26 Oral
- EMNLP’25 FindingsFC-Attack: Jailbreaking Multimodal Large Language Models via Auto-Generated FlowchartsIn EMNLP Findings 2025
- ACL’25 MainAre We in the AI-Generated Text World Already? Quantifying and Monitoring AIGT on Social MediaIn ACL Main 2025
- SocialMeta’24AdSpectorX: A Multimodal Expert Spector for Covert Advertising Detection on Chinese Social MediaIn SocialMeta Workshop 2024 — Best Paper Award