English
Related papers

Related papers: Before the Last Token: Diagnosing Final-Token Safe…

200 papers

Large Language Models (LLMs) have gained significant attention but also raised concerns due to the risk of misuse. Jailbreak prompts, a popular type of adversarial attack towards LLMs, have appeared and constantly evolved to breach the…

Human-Computer Interaction · Computer Science 2024-07-04 Zhihua Jin , Shiyi Liu , Haotian Li , Xun Zhao , Huamin Qu

Given the power of vision transformers, a new learning paradigm, pre-training and then prompting, makes it more efficient and effective to address downstream visual recognition tasks. In this paper, we identify a novel security threat…

Computer Vision and Pattern Recognition · Computer Science 2024-05-20 Sheng Yang , Jiawang Bai , Kuofeng Gao , Yong Yang , Yiming Li , Shu-tao Xia

Despite the impressive capabilities of Large Language Models (LLMs) in various tasks, their vulnerability to unsafe prompts remains a critical issue. These prompts can lead LLMs to generate responses on illegal or sensitive topics, posing a…

Computation and Language · Computer Science 2024-07-10 Jinseok Kim , Jaewon Jung , Sangyeop Kim , Sohyung Park , Sungzoon Cho

Text-to-image (T2I) models can generate not-safe-for-work (NSFW) content, motivating multi-stage safety pipelines with both text and image filters. Newer LLM-based filters detect latent intent beyond keywords, making token-level…

Machine Learning · Computer Science 2026-05-26 Zixuan Chen , Hao Lin , Ke Xu , Xinghao Jiang , Tanfeng Sun

Despite recent advances, Large Language Models remain vulnerable to jailbreak attacks that bypass alignment safeguards and elicit harmful outputs. While prior research has proposed various attack strategies differing in human readability…

Computation and Language · Computer Science 2025-10-28 Havva Alizadeh Noughabi , Julien Serbanescu , Fattane Zarrinkalam , Ali Dehghantanha

Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks, yet they remain vulnerable to adversarial manipulations such as jailbreaking via prompt injection attacks. These attacks bypass safety mechanisms…

Machine Learning · Computer Science 2025-07-08 Xin Wei Chia , Swee Liang Wong , Jonathan Pan

As large language models (LLMs) are increasingly deployed, ensuring their safe use is paramount. Jailbreaking, adversarial prompts that bypass model alignment to trigger harmful outputs, present significant risks, with existing studies…

Cryptography and Security · Computer Science 2026-01-01 Yuan Xin , Dingfan Chen , Linyi Yang , Michael Backes , Xiao Zhang

LLM-based coding assistants are seeing rapid adoption, offering substantial gains in developer productivity. As organizations increasingly ship code these agents produce, the security of that code becomes critical. Prior work has shown that…

Cryptography and Security · Computer Science 2026-05-29 Alexander Sternfeld , Andrei Kucharavy , Ljiljana Dolamic

Large language models (LLMs) are known to be vulnerable to jailbreak attacks, which typically rely on carefully designed prompts containing explicit semantic structure. These attacks generally operate by fixing an adversarial instruction…

Machine Learning · Computer Science 2026-05-07 Marco Rando , Samuel Vaiter

Jailbreaking in Large Language Models (LLMs) threatens their safe use in sensitive domains like education by allowing users to bypass ethical safeguards. This study focuses on detecting jailbreaks in 2-Sigma, a clinical education platform…

LLM-integrated applications are vulnerable to prompt injection attacks, where an attacker contaminates the input to inject malicious instructions, causing the LLM to follow the attacker's intent instead of the original user's. Existing…

Cryptography and Security · Computer Science 2026-01-27 Wei Zou , Yupei Liu , Yanting Wang , Ying Chen , Neil Gong , Jinyuan Jia

Large (vision-)language models exhibit remarkable capability but remain highly susceptible to jailbreaking. Existing safety training approaches aim to have the model learn a refusal boundary between safe and unsafe, based on the user's…

Cryptography and Security · Computer Science 2026-04-28 Xinhe Wang , Katia Sycara , Yaqi Xie

Ensuring the safety and alignment of Large Language Models is a significant challenge with their growing integration into critical applications and societal functions. While prior research has primarily focused on jailbreak attacks, less…

Machine Learning · Computer Science 2026-04-28 Jiawei Chen , Zhengwei Fang , Yu Tian , Jiawei Du , Chao Yu , Zhaoxia Yin , Hang Su

Large Language Models (LLMs) are increasingly exposed to adaptive jailbreaking, particularly in high-stakes Chemical, Biological, Radiological, and Nuclear (CBRN) domains. Although streaming probes enable real-time monitoring, they still…

Computation and Language · Computer Science 2026-04-17 Xuanli He , Bilgehan Sel , Faizan Ali , Jenny Bao , Hoagy Cunningham , Jerry Wei

Recent explorations with commercial Large Language Models (LLMs) have shown that non-expert users can jailbreak LLMs by simply manipulating their prompts; resulting in degenerate output behavior, privacy and security breaches, offensive…

Computation and Language · Computer Science 2024-03-28 Abhinav Rao , Sachin Vashistha , Atharva Naik , Somak Aditya , Monojit Choudhury

Reasoning language models improve performance on complex tasks by generating long chains of thought (CoTs), but this process can also increase harmful outputs in adversarial settings. In this work, we ask whether the long CoTs can be…

Computation and Language · Computer Science 2025-10-08 Yik Siu Chan , Zheng-Xin Yong , Stephen H. Bach

Prompt injection and jailbreaking attacks pose persistent security challenges to large language model (LLM)-based systems. We present PromptScreen, an efficient and systematically evaluated defense architecture that mitigates these threats…

Cryptography and Security · Computer Science 2026-01-12 Akshaj Prashanth Rao , Advait Singh , Saumya Kumaar Saksena , Dhruv Kumar

Activation probes are attractive monitors for AI systems due to low cost and latency, but their real-world robustness remains underexplored. We ask: What failure modes arise under realistic, black-box adversarial pressure, and how can we…

Machine Learning · Computer Science 2025-11-04 Phil Blandfort , Robert Graham

Toxicity detection is crucial for maintaining the peace of the society. While existing methods perform well on normal toxic contents or those generated by specific perturbation methods, they are vulnerable to evolving perturbation patterns.…

Cryptography and Security · Computer Science 2025-03-05 Hankun Kang , Jianhao Chen , Yongqi Li , Xin Miao , Mayi Xu , Ming Zhong , Yuanyuan Zhu , Tieyun Qian

Safety mechanisms for large language models (LLMs) remain predominantly English-centric, creating systematic vulnerabilities in multilingual deployment. Prior work shows that translating malicious prompts into other languages can…

Computation and Language · Computer Science 2026-04-29 Shirin Alanova , Bogdan Minko , Sabrina Sadiekh , Evgeniy Kokuykin