Zongmin Zhang

Ph.D. Student · Trustworthy AI

Zongmin Zhang on an offshore fishing trip

HKUST (Guangzhou)

Guangzhou, China

Hello! I am a 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 lies in Trustworthy AI, with a focus on the security and privacy of multimodal foundation models. I study backdoor and jailbreak attacks, adversarial robustness, and content authenticity across vision, language, and audio 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.

AI Security & Privacy

Backdoors, jailbreaks, adversarial attacks, and defenses for modern AI systems.

Multimodal Foundation Models

Security and safety of vision segmentation models, LLMs, and VLLMs.

Content Authenticity

Image watermarking, synthetic media detection, and trustworthy audio-visual content.

News

2026.08 Our paper MARS was accepted to NDSS 2027.
2025.11 Our paper 6DAttack was accepted to AAAI 2026 as an Oral presentation.
2025.09 Our paper CHASM was accepted to NeurIPS 2025.
2025.09 Our paper FC-Attack was accepted to EMNLP 2025 Findings.
2025.05 Our paper on AI-generated text in social media was accepted to ACL 2025 Main.

Selected Publications

* indicates equal contribution.

  1. NDSS NDSS’27
    Escaping the Linearity Trap: Manifold Detours for Black-Box Adversarial Attacks on Singing Audio Deepfake Detection
    Yifan Liao, Yule Liu, Zhen Sun, Zongmin Zhang, Yupeng He, Jiaheng Wei, Xinhu Zheng, and Xinlei He
    In NDSS 2027
    arXiv
  2. arXiv arXiv’25
    Backdoor Attacks on Prompt-Driven Video Segmentation Foundation Models
    Zongmin Zhang*, Zhen Sun*, Yifan Liao, Wenhan Dong, Xinlei He, Xingshuo Han, Shengmin Xu, and Xinyi Huang
    arXiv preprint Dec 2025
    arXiv
  3. arXiv arXiv’25
    To Survive, I Must Defect: Jailbreaking LLMs via the Game-Theory Scenarios
    Zhen Sun*, Zongmin Zhang*, Deqi Liang, Han Sun, Yule Liu, Yun Shen, Xiangshan Gao, Yilong Yang, Shuai Liu, Yutao Yue, and Xinlei He
    arXiv preprint Nov 2025
    arXiv
  4. AAAI AAAI’26 Oral
    6DAttack: Backdoor Attacks in the 6DoF Pose Estimation
    Jihui Guo, Zongmin Zhang, Zhen Sun, Yuhao Yang, Jinlin Wu, Fu Zhang, and Xinlei He
    In AAAI Oral 2026
    arXiv
  5. NeurIPS NeurIPS’25
    CHASM: Unveiling Covert Advertisements on Chinese Social Media
    Jingyi Zheng, Tianyi Hu, Yule Liu, Zhen Sun, Zongmin Zhang, Zifan Peng, Wenhan Dong, and Xinlei He
    In NeurIPS 2025
    arXiv Code Website
  6. EMNLP EMNLP’25 Findings
    FC-Attack: Jailbreaking Multimodal Large Language Models via Auto-Generated Flowcharts
    Ziyi Zhang, Zhen Sun, Zongmin Zhang, Jihui Guo, and Xinlei He
    In EMNLP Findings 2025
    arXiv
  7. ACL ACL’25 Main
    Are We in the AI-Generated Text World Already? Quantifying and Monitoring AIGT on Social Media
    Zhen Sun*, Zongmin Zhang*, Xinyue Shen, Ziyi Zhang, Yule Liu, Michael Backes, Yang Zhang, and Xinlei He
    In ACL Main 2025
    arXiv
  8. SocialMeta SocialMeta’24
    AdSpectorX: A Multimodal Expert Spector for Covert Advertising Detection on Chinese Social Media
    Zongmin Zhang, Yujie Han, Zhou Zhang, Yule Liu, Jingyi Zheng, and Zhen Sun
    In SocialMeta Workshop 2024 — Best Paper Award
    HTML