袁增辉

袁增辉，杭州电子科技大学计算机学院特聘副研究员，隶属于云技术团队。2025年博士毕业于华中科技大学网络空间安全学院，师从周潘教授。研究方向包括人工智能安全、大模型及其应用、智能体安全等，于CCS、NDSS、NeurIPS、CVPR、ACL等国际会议与期刊上发表论文10篇。担任NeurIPS、CVPR、ECCV、AAAI、IEEE TPAMI、IEEE TDSC等会议与期刊的审稿人。获2024年“华为杯”全国研究生网安创新大赛一等奖。

• 谷歌学术主页：https://scholar.google.com/citations?user=YhrXMbkAAAAJ&hl=zh-CN

• 欢迎有兴趣做科研的本科生、研究生加入团队，一起冲击顶会论文！
  

• 联系邮箱：zenghuiyuan@hdu.edu.cn

• 2015.09-2019.06，四川大学，学士

• 2019.09-2021.06，华中科技大学，硕士

• 2021.09-2025.09，华中科技大学，博士
  

• 2025年11月-至今 杭州电子科技大学计算机学院 特聘副研究员

代表性成果（*表示共同一作）：

• Zhi Luo, Zenghui Yuan, Wenqi Wei, etc., An Image Is Worth Ten Thousand Words: Verbose-Text Induction Attacks on VLMs, The 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2026. (CCF-B)

• Guiyao Tie*, Zenghui Yuan*, Zeli Zhao*, etc., Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs, The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025. (CCF-A)

• Hai Yan, Haijian Ma, Xiaowen Cai, Daizong Liu, Zenghui Yuan, etc., Fit the Distribution: Cross-Image/Prompt Adversarial Attacks on Multimodal Large Language Models, The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025. (CCF-A)

• Jiawen Shi, Zenghui Yuan, Guiyao Tie, etc., Prompt Injection Attack to Tool Selection in LLM Agents, ISOC Network and Distributed System Security Symposium (NDSS), 2026. (CCF-A)

• Zenghui Yuan, Yangming Xu, Jiawen Shi, etc., Merge Hijacking: Backdoor Attacks to Model Merging of Large Language Models, Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL), 2025. (CCF-A)

• Zenghui Yuan, Jiawen Shi, Pan Zhou, etc., Badtoken: Token-level backdoor attacks to multi-modal large language models, Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR), 2025. (CCF-A)

• Jiawen Shi*, Zenghui Yuan*, Yinuo Liu, etc., Optimization-based prompt injection attack to llm-as-a-judge, Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security (CCS), 2024. (CCF-A)

• Zenghui Yuan, Pan Zhou, Kai Zou, etc., You are catching my attention: Are vision transformers bad learners under backdoor attacks?, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023. (CCF-A)

• Cheng Wang*, Zenghui Yuan*, Pan Zhou, etc., The security and privacy of mobile-edge computing: An artificial intelligence perspective, IEEE Internet of Things Journal (IoT Journal). (SCI二区)