TrojanStego: Your Language Model Can Secretly Be A Steganographic Privacy Leaking Agent
Authors: Dominik Meier, Jan Philip Wahle, Paul Röttger, Terry Ruas, Bela Gipp | Published: 2025-05-26
Prompt Injection
What Really Matters in Many-Shot Attacks? An Empirical Study of Long-Context Vulnerabilities in LLMs
Authors: Sangyeop Kim, Yohan Lee, Yongwoo Song, Kimin Lee | Published: 2025-05-26
Invisible Prompts, Visible Threats: Malicious Font Injection in External Resources for Large Language Models
Authors: Junjie Xiong, Changjia Zhu, Shuhang Lin, Chong Zhang, Yongfeng Zhang, Yao Liu, Lingyao Li | Published: 2025-05-22
CAIN: Hijacking LLM-Humans Conversations via a Two-Stage Malicious System Prompt Generation and Refining Framework
Authors: Viet Pham, Thai Le | Published: 2025-05-22
When Safety Detectors Aren’t Enough: A Stealthy and Effective Jailbreak Attack on LLMs via Steganographic Techniques
Authors: Jianing Geng, Biao Yi, Zekun Fei, Tongxi Wu, Lihai Nie, Zheli Liu | Published: 2025-05-22
BitHydra: Towards Bit-flip Inference Cost Attack against Large Language Models
Authors: Xiaobei Yan, Yiming Li, Zhaoxin Fan, Han Qiu, Tianwei Zhang | Published: 2025-05-22
Finetuning-Activated Backdoors in LLMs
Authors: Thibaud Gloaguen, Mark Vero, Robin Staab, Martin Vechev | Published: 2025-05-22
DuFFin: A Dual-Level Fingerprinting Framework for LLMs IP Protection
Authors: Yuliang Yan, Haochun Tang, Shuo Yan, Enyan Dai | Published: 2025-05-22
Alignment Under Pressure: The Case for Informed Adversaries When Evaluating LLM Defenses
Authors: Xiaoxue Yang, Bozhidar Stevanoski, Matthieu Meeus, Yves-Alexandre de Montjoye | Published: 2025-05-21
sudoLLM : On Multi-role Alignment of Language Models
Authors: Soumadeep Saha, Akshay Chaturvedi, Joy Mahapatra, Utpal Garain | Published: 2025-05-20