English
Related papers

Related papers: Best-of-N Jailbreaking

200 papers

Recent work showed Best-of-N (BoN) jailbreaking using repeated use of random augmentations (such as capitalization, punctuation, etc) is effective against all major large language models (LLMs). We have found that $100\%$ of the BoN paper's…

Cryptography and Security · Computer Science 2025-02-04 Stuart Armstrong , Matija Franklin , Connor Stevens , Rebecca Gorman

Best-of-N (BoN) is a popular and effective algorithm for aligning language models to human preferences. The algorithm works as follows: at inference time, N samples are drawn from the language model, and the sample with the highest reward,…

Computation and Language · Computer Science 2025-03-05 Afra Amini , Tim Vieira , Elliott Ash , Ryan Cotterell

Safety alignment mechanism are essential for preventing large language models (LLMs) from generating harmful information or unethical content. However, cleverly crafted prompts can bypass these safety measures without accessing the model's…

Computation and Language · Computer Science 2025-01-31 Sunbowen Lee , Shiwen Ni , Chi Wei , Shuaimin Li , Liyang Fan , Ahmadreza Argha , Hamid Alinejad-Rokny , Ruifeng Xu , Yicheng Gong , Min Yang

With the rapid advancement of large language models (LLMs), ensuring their safe use becomes increasingly critical. Fine-tuning is a widely used method for adapting models to downstream tasks, yet it is vulnerable to jailbreak attacks.…

Cryptography and Security · Computer Science 2025-10-10 Xiangfang Li , Yu Wang , Bo Li

Best-of-N (BoN) sampling with a reward model has been shown to be an effective strategy for aligning Large Language Models (LLMs) with human preferences at the time of decoding. BoN sampling is susceptible to a problem known as reward…

Computation and Language · Computer Science 2025-02-19 Yuki Ichihara , Yuu Jinnai , Tetsuro Morimura , Kaito Ariu , Kenshi Abe , Mitsuki Sakamoto , Eiji Uchibe

The rapid advancement of Large Language Models (LLMs) has introduced significant challenges in moderating user-model interactions. While LLMs demonstrate remarkable capabilities, they remain vulnerable to adversarial attacks, particularly…

Cryptography and Security · Computer Science 2025-02-14 Ivan Bakulin , Ilia Kopanichuk , Iaroslav Bespalov , Nikita Radchenko , Vladimir Shaposhnikov , Dmitry Dylov , Ivan Oseledets

Sampling multiple outputs from a Large Language Model (LLM) and selecting the most frequent (Self-consistency) or highest-scoring (Best-of-N) candidate is a popular approach to achieve higher accuracy in tasks with discrete final answers.…

Machine Learning · Computer Science 2025-11-25 Amin Rakhsha , Kanika Madan , Tianyu Zhang , Amir-massoud Farahmand , Amir Khasahmadi

AI systems are rapidly advancing in capability, and frontier model developers broadly acknowledge the need for safeguards against serious misuse. However, this paper demonstrates that fine-tuning, whether via open weights or closed…

Cryptography and Security · Computer Science 2025-09-23 Brendan Murphy , Dillon Bowen , Shahrad Mohammadzadeh , Tom Tseng , Julius Broomfield , Adam Gleave , Kellin Pelrine

Best-of-N (BoN) sampling with a reward model has been shown to be an effective strategy for aligning Large Language Models (LLMs) to human preferences at the time of decoding. BoN sampling is susceptible to a problem known as reward hacking…

Computation and Language · Computer Science 2025-01-30 Yuu Jinnai , Tetsuro Morimura , Kaito Ariu , Kenshi Abe

Vision-language models (VLMs) have become central to tasks such as visual question answering, image captioning, and text-to-image generation. However, their outputs are highly sensitive to prompt variations, which can reveal vulnerabilities…

Computer Vision and Pattern Recognition · Computer Science 2026-02-02 Aarush Noheria , Yuguang Yao

We study a new vulnerability in commercial-scale safety-aligned large language models (LLMs): their refusal to generate harmful responses can be broken by flipping only a few bits in model parameters. Our attack jailbreaks billion-parameter…

Cryptography and Security · Computer Science 2025-10-06 Zachary Coalson , Jeonghyun Woo , Chris S. Lin , Joyce Qu , Yu Sun , Shiyang Chen , Lishan Yang , Gururaj Saileshwar , Prashant Nair , Bo Fang , Sanghyun Hong

Despite advances in AI alignment, large language models (LLMs) remain vulnerable to adversarial attacks or jailbreaking, in which adversaries can modify prompts to induce unwanted behavior. While some defenses have been proposed, they have…

Machine Learning · Computer Science 2024-11-11 Andy Zhou , Bo Li , Haohan Wang

While large language models (LLMs) exhibit remarkable capabilities across various tasks, they encounter potential security risks such as jailbreak attacks, which exploit vulnerabilities to bypass security measures and generate harmful…

Cryptography and Security · Computer Science 2024-11-28 Xinyuan Wang , Victor Shea-Jay Huang , Renmiao Chen , Hao Wang , Chengwei Pan , Lei Sha , Minlie Huang

Best-of-$n$ (BoN) sampling is a practical approach for aligning language model outputs with human preferences without expensive fine-tuning. BoN sampling is performed by generating $n$ responses to a prompt and then selecting the sample…

Information Theory · Computer Science 2025-05-07 Claudio Mayrink Verdun , Alex Oesterling , Himabindu Lakkaraju , Flavio P. Calmon

Large Language Models face security threats from jailbreak attacks. Existing research has predominantly focused on prompt-level attacks while largely ignoring the underexplored attack surface of user-controlled response prefilling. This…

Cryptography and Security · Computer Science 2025-08-27 Yakai Li , Jiekang Hu , Weiduan Sang , Luping Ma , Dongsheng Nie , Weijuan Zhang , Aimin Yu , Yi Su , Qingjia Huang , Qihang Zhou

Large language models remain vulnerable to jailbreak attacks, yet we still lack a systematic understanding of how jailbreak success scales with attacker effort across methods, model families, and harm types. We initiate a scaling-law…

Machine Learning · Computer Science 2026-03-20 Xiangwen Wang , Ananth Balashankar , Varun Chandrasekaran

Large Language Models (LLMs) can be used to red team other models (e.g. jailbreaking) to elicit harmful contents. While prior works commonly employ open-weight models or private uncensored models for doing jailbreaking, as the…

Large language models (LLMs) have seen widespread applications across various domains, yet remain vulnerable to adversarial prompt injections. While most existing research on jailbreak attacks and hallucination phenomena has focused…

Computation and Language · Computer Science 2025-05-20 Linghan Huang , Haolin Jin , Zhaoge Bi , Pengyue Yang , Peizhou Zhao , Taozhao Chen , Xiongfei Wu , Lei Ma , Huaming Chen

Large language model (LLM) safety is a critical issue, with numerous studies employing red team testing to enhance model security. Among these, jailbreak methods explore potential vulnerabilities by crafting malicious prompts that induce…

Computation and Language · Computer Science 2025-03-07 Honglin Mu , Han He , Yuxin Zhou , Yunlong Feng , Yang Xu , Libo Qin , Xiaoming Shi , Zeming Liu , Xudong Han , Qi Shi , Qingfu Zhu , Wanxiang Che

In deployment and application, large language models (LLMs) typically undergo safety alignment to prevent illegal and unethical outputs. However, the continuous advancement of jailbreak attack techniques, designed to bypass safety…

Cryptography and Security · Computer Science 2025-09-05 Chuhan Zhang , Ye Zhang , Bowen Shi , Yuyou Gan , Tianyu Du , Shouling Ji , Dazhan Deng , Yingcai Wu
‹ Prev 1 2 3 10 Next ›