姓名：陈康杰

性别：男

出生年月：1992年08月

学历/学位：博士研究生/工学博士

职称：教授

所在院系/专业：网络安全学院

2011.09-2015.06   电子科技大学 自动化 学士

2016.09-2019.06   天津大学 计算机技术 硕士

2020.08-2025.02   新加坡南洋理工大学 计算机科学与技术 博士

可信人工智能，智能漏洞分析，具身智能安全。

研究方向为可信人工智能和智能漏洞分析，主要聚焦于大模型智能体安全以及自动化漏洞挖掘分析等方向的研究。

近年来，围绕大语言模型与多模态大模型的安全评测、自动化红队测试、后门攻击与检测、模型安全对齐等问题开展了系统研究。相关成果发表于 ICML、ICLR、NeurIPS、USENIX Security、CCS、NDSS、TIFS 、TDSC等人工智能与安全领域重要会议和期刊，并获得了ICLR Oral、CVPR Oral及NeurIPS Spotlight等论文奖。

[1] TRUST-VLM: Thorough Red-teaming for Uncovering Safety Threats in Vision-Language Models. Kangjie Chen, Muyang Li, Guanlin Li, Shudong Zhang, Shangwei Guo, Tianwei Zhang. ICML 2025. (CCF A类会议)

[2] Clean-image Backdoor: Attacking Multi-label Models with Poisoned Labels Only. Kangjie Chen, Xiaoxuan Lou, Guowen Xu, Jiwei Li, Tianwei Zhang. ICLR  2023, Oral. (CCF A类会议)

[3] BadPre: Task-agnostic Backdoor Attacks to Pre-trained NLP Foundation Models. Kangjie Chen, Yuxian Meng, Xiaofei Sun, Shangwei Guo, Tianwei Zhang, Jiwei Li, Chun Fan. ICLR 2022. (CCF A类会议)

[4] USD: NSFW Content Detection for Text-to-Image Models via Scene Graph. Yuyang Zhang*, Kangjie Chen*, Xudong Jiang, Jiahui Wen, Yihui Jin, Ziyou Liang, Yihao Huang, Run Wang, Lina Wang.  USENIX Security 2025. (CCF A类会议)

[5] ART: Automatic Red-teaming for Text-to-Image Models to Protect Benign Users. Guanlin Li, Kangjie Chen#, Shudong Zhang, Jie Zhang, Tianwei Zhang. NeurIPS 2024. (CCF A类会议)

[6] Unmasking Backdoors: An Explainable Defense via Gradient-Attention Anomaly Scoring for Pre-trained Language Models. Anindya Sundar Das, Kangjie Chen#, Monowar Bhuyan. ICLR 2026. (CCF A类论文)

[7] When Search Goes Wrong: Red-Teaming Web-Augmented Large Language Models. Haoran Ou, Kangjie Chen#, Xingshuo Han, Gelei Deng, Jie Zhang, Han Qiu, Tianwei Zhang. ICML 2026. (CCF-A类论文)

1. 入选国家级青年人才，2025年;

2. Clean-image Backdoor: Attacking Multi-label Models with Poisoned Labels Only. ICLR 2023, Oral;

3. Transstratal Adversarial Attack: Compromising Multi-Layered Defenses in Text-to-Image Models. NeurIPS 2025, Spotlight;

4. A Unified Anomaly Detection Methodology for Lane-Following of Autonomous Driving Systems. ISPA 2021, Most Innovative Paper Award.

办公地址：天津大学北洋园校区55教学楼B110

电子邮箱：kangjie_chen@tju.edu.cn