The Resurgence of GCG Adversarial Attacks on Large Language Models.

Yuting Tan 0008 HydroX AI, USA Yuting Tan Yuting Tan 0001 Central China Normal University, Wuhan, China Yuting Tan 0002 Communication University of China, Beijing, China https://orcid.org/0009-0006-8859-481X Yuting Tan 0003 Dalian Maritime University, Dalian, China https://orcid.org/0000-0002-3605-1307 Yuting Tan 0004 Donghua University, Shanghai, China Yuting Tan 0005 Vanderbilt University, Department of Molecular Physiology and Biophysics, Nashville, TN, USA Yuting Tan 0006 Microsoft, Redmond, WA, USA Yuting Tan 0007 Guangdong University of Technology, School of Management, Guangzhou, China https://orcid.org/0009-0004-9951-3398 Yuting Tan 0009 Chongqing University of Posts and Telecommunications, School of Electronic Science and Engineering, China https://orcid.org/0009-0001-0417-7765 Yuting Tan 0010 Education University of Hong Kong, Department of Psychology, Hong Kong Beijing Normal University, Collaborative Innovation Center of Assessment Toward Basic Education Quality, China

Yuting Tan 0008 Xuying Li Zhuo Li Huizhen Shu Peikang Hu The Resurgence of GCG Adversarial Attacks on Large Language Models. 2025 September abs/2509.00391 CoRR https://doi.org/10.48550/arXiv.2509.00391 db/journals/corr/corr2509.html#abs-2509-00391 streams/journals/corr

Yuting Tan 0008 Yi Huang Zhuo Li The 'Sure' Trap: Multi-Scale Poisoning Analysis of Stealthy Compliance-Only Backdoors in Fine-Tuned Large Language Models. 2025 November abs/2511.12414 CoRR https://doi.org/10.48550/arXiv.2511.12414 db/journals/corr/corr2511.html#abs-2511-12414 streams/journals/corr

Peikang Hu Yi Huang Xuying Li Zhuo Li Huizhen Shu