I obtained my Computer Science Ph.D. degree under the supervision of Prof. Tianwei Zhang in S-Lab, Nanyang Technological University, Singapore. Before that, I received my B.Eng. degree in Information Security, Mathematics from Shandong University, China.

My research focuses on critical aspects of artificial intelligence and machine learning security. Specifically, I investigate adversarial attacks and defenses, examining how malicious entities can exploit vulnerabilities in AI systems and developing robust strategies to mitigate these threats. I also explore the security of AI-generated content (AIGC), ensuring that generated outputs are safe and reliable. My work in red-teaming for models involves simulating adversarial and non-adversarial scenarios to test and improve the resilience of AI systems. Furthermore, I explore model intellectual property protection, devising methods to safeguard proprietary AI models from unauthorized access and misuse.

I am currently studying to build an Agent to help model developers find safety risks (jailbreak, adversarial examples, and so on) in generative models, such as LLM, VLM, and others. Contact me if you are interested and want cooperation. Note only for research studying.

Research Interests

• Deep Learning
• Computer Vision
• Adversarial Attack and Defense
• Backdoor Attack and Data Poison
• Security of Large Generative Models

Researches (A complete list can be found in my Google Scholar)

• Picky LLMs and Unreliable RMs: An Empirical Study on Safety Alignment after Instruction Tuning [pdf][code]

  Guanlin Li, Kangjie Chen, Shangwei Guo, Jie Zhang, Han Qiu, Chao Zhang, Guoyin Wang, Tianwei Zhang, Jiwei Li

  arXiv, 2025

• ART: Automatic Red-teaming for Text-to-Image Models to Protect Benign Users [pdf][code]

  Guanlin Li, Kangjie Chen, Shudong Zhang, Jie Zhang, Tianwei Zhang

  NeurIPS, 2024

• Fingerprinting Image-to-Image Generative Adversarial Networks [pdf]

  Guanlin Li, Guowen Xu, Han Qiu, Shangwei Guo, Run Wang, Jiwei Li, Tianwei Zhang, Rongxing Lu

  EuroS&P, 2024

• PRIME: Protect Your Videos From Malicious Editing [pdf][code]

  Guanlin Li, Shuai Yang, Jie Zhang, Tianwei Zhang

  arXiv, 2024

• Warfare:Breaking the Watermark Protection of AI-Generated Content [pdf][code]

  Guanlin Li, Yifei Chen, Jie Zhang, Shangwei Guo, Han Qiu, Guoyin Wang, Jiwei Li, Tianwei Zhang

  arXiv, 2023

• Singular Regularization with Information Bottleneck Improves Model’s Adversarial Robustness [pdf]

  Guanlin Li, Naishan Zheng, Man Zhou, Jie Zhang, Tianwei Zhang

  arXiv, 2023

• Rethinking Adversarial Training with Neural Tangent Kernel [pdf]

  Guanlin Li, Han Qiu, Shangwei Guo, Jiwei Li, Tianwei Zhang

  arXiv, 2023

• Alleviating the Effect of Data Imbalance on Adversarial Training [pdf] [code]

  Guanlin Li, Guowen Xu, Tianwei Zhang

  arXiv, 2023

• Omnipotent Adversarial Training in the Wild [pdf] [code]

  Guanlin Li, Kangjie Chen, Yuan Xu, Han Qiu, Tianwei Zhang

  arXiv, 2023

• Extracting Robust Models with Uncertain Examples [pdf] [code]

  Guanlin Li, Guowen Xu, Shangwei Guo, Han Qiu, Jiwei Li, Tianwei Zhang

  ICLR, 2023

• Secure Decentralized Image Classification with Multiparty Homomorphic Encryption [pdf]

  Guowen Xu, Guanlin Li, Shangwei Guo, Tianwei Zhang, Hongwei Li

  IEEE Transactions on Circuits and Systems for Video Technology, 2023

• A Benchmark of Long-tailed Instance Segmentation with Noisy Labels [pdf] [code]

  Guanlin Li, Guowen Xu, Tianwei Zhang

  arXiv, 2022

• Improving Adversarial Robustness of 3D Point Cloud Classification Models [pdf] [code]

  Guanlin Li, Guowen Xu, Han Qiu, Ruan He, Jiwei Li, Tianwei Zhang

  ECCV, 2022

• Enhancing intrinsic adversarial robustness via feature pyramid decoder [pdf] [code]

  Guanlin Li, Shuya Ding, Jun Luo, Chang Liu

  CVPR, 2020

• Scnet: A neural network for automated side-channel attack [pdf] [code]

  Guanlin Li, Chang Liu, Han Yu, Yanhong Fan, Libang Zhang, Zongyue Wang, Meiqin Wang

  arXiv, 2020

Professional Services

Conference Reviewer for ICML, NeurIPS, ICLR, ECCV, ICCV and CVPR