Gelei Deng ☕️
Gelei Deng
NTU
OpenAI Red Team

    About Me

    My research focuses on AI safety and AI security. I am particularly interested in leveraging AI and automated systems to attack AI and cyber systems autonomously, enabling scalable and intelligent security testing. I received my PhD from Nanyang Technological University, advised by Prof. Tianwei Zhang and Prof. Yang Liu.

    Download CV
    Interests
    • AI Security and Safety
    • Large Language Models
    • Penetration Testing
    • Blockchain Security
    • System Security
    Education

    • PhD in Computer Science

      Nanyang Technological University

    • B.E. Electrical Engineering

      Singapore University of Technology and Design

    Featured Research
    PentestGPT: Evaluating and Harnessing Large Language Models for Automated Penetration Testing
    MASTERKEY: Automated Jailbreaking of Large Language Model Chatbots
    Selected Projects

    Research artifacts, open-source systems, and security testing frameworks from my recent work.

    Large Language Models

    PentestGPT

    An LLM-empowered automatic penetration testing framework with 14k+ GitHub stars and 2.4k+ forks. PentestGPT is designed to automate penetration testing by leveraging the domain knowledge inherent in Large Language Models. It features a three-module architecture (Reasoning, Generation, and Parsing) that emulates human penetration testing workflows. Key Features: Multi-module agent design for reasoning, generation, and parsing Integration with multiple LLM backends and real-world security workflows Evaluation on CTF challenges and practical penetration testing targets 228.6% task completion improvement over baseline GPT models Recognition: Distinguished Artifact Award at USENIX Security 2024 Widely used open-source security research artifact with active community adoption

    Recent Publications
    Gelei Deng, Yi Liu, Yuekang Li, Ruozhao Yang, Xiaofei Xie, Jie Zhang, Han Qiu, Tianwei Zhang (2026). What Makes a Good LLM Agent for Real-world Penetration Testing?. arXiv 2026.
    PDF Project DOI arXiv
    Yi Liu, Weizhe Wang, Ruitao Feng, Yao Zhang, Guangquan Xu, Gelei Deng, Yuekang Li, Leo Zhang (2026). Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale. arXiv 2026.
    PDF DOI arXiv
    Jingnan Zheng, Xiangtian Ji, Yijun Lu, Chenhang Cui, Weixiang Zhao, Gelei Deng, Zhenkai Liang, An Zhang, Tat-Seng Chua (2025). RSafe: Incentivizing Proactive Reasoning to Build Robust and Adaptive LLM Safeguards. NeurIPS 2025.
    PDF DOI arXiv
    Chenhang Cui, Gelei Deng, An Zhang, Jingnan Zheng, Yicong Li, Lianli Gao, Tianwei Zhang, Tat-Seng Chua (2025). Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models. NeurIPS 2025.
    PDF Code Project DOI arXiv Code
    Yuan Xu, Gelei Deng, Guanlin Li, Xingshuo Han, Shangwei Guo, Tianwei Zhang (2025). Controllable Spoofing Attacks on Visual SLAM in Robotic Vehicles. ACSAC 2025.
    ACSAC
    Cheng Wang, Gelei Deng, Xianglin Yang, Han Qiu, Tianwei Zhang (2025). When Audio and Text Disagree: Revealing Text Bias in Large Audio-Language Models. EMNLP 2025.
    PDF EMNLP arXiv
    Gelei Deng, Haoran Ou, Yi Liu, Jie Zhang, Tianwei Zhang, Yang Liu (2025). Oedipus: LLM-enchanced Reasoning CAPTCHA Solver. CCS 2025.
    PDF arXiv PDF
    Yuan Xu, Gelei Deng, Tianwei Zhang (2025). Detecting Perception-Based Attacks using Visual Odometry: Inconsistency Modeling and Checking on Robotic States. ICRA 2025.
    Ziqi Ding, Gelei Deng, Yi Liu, Junchen Ding, Jieshan Chen, Yulei Sui, Yuekang Li (2025). IllusionCAPTCHA: A CAPTCHA based on Visual Illusion. WWW 2025.
    PDF DOI arXiv
    Weisong Sun, Yiming Miao, Yuekang Li, Hongyu Zhang, Chunrong Fang, Yi Liu, Gelei Deng, Yang Liu, Zhenyu Chen (2025). Source Code Summarization in the Era of Large Language Models. ICSE 2025.
    PDF arXiv
    Yi Liu, Junzhe Yu, Huijia Sun, Ling Shi, Gelei Deng, Yuqi Chen, Yang Liu (2024). Efficient Detection of Toxic Prompts in Large Language Models. ASE 2024.
    PDF DOI arXiv
    Kunsheng Tang, Wenbo Zhou, Jie Zhang, Aishan Liu, Gelei Deng, Shuai Li, Peigui Qi, Weiming Zhang, Tianwei Zhang, Nenghai Yu (2024). GenderCARE: A Comprehensive Framework for Assessing and Reducing Gender Bias in Large Language Models. CCS 2024.
    PDF Code DOI arXiv Code
    See all publications