paper-with-me

Papers

Exploring Jailbreak Attacks on LLMs through Intent Concealment and Diversion

2025-05-20 · Tiehan Cui, Yanxu Mao, Peipei Liu, Congying Liu, Datao You

Although large language models (LLMs) have achieved remarkable advancements, their security remains a pressing concern. One major threat is jailbreak attacks, where adversarial prompts bypass model safeguards to generate harmful or objectionable content. Researchers study jailbreak attacks to understand security and robustness of LLMs. However, existing jailbreak attack methods face two main challenges: (1) an excessive number of iterative queries, and (2) poor generalization across models. In addition, recent jailbreak evaluation datasets focus primarily on question-answering scenarios, lacking attention to text generation tasks that require accurate regeneration of toxic content. To tackle these challenges, we propose two contributions: (1) ICE, a novel black-box jailbreak method that employs Intent Concealment and divErsion to effectively circumvent security constraints. ICE achieves high attack success rates (ASR) with a single query, significantly improving efficiency and transferability across different models. (2) BiSceneEval, a comprehensive dataset designed for assessing LLM robustness in question-answering and text-generation tasks. Experimental results demonstrate that ICE outperforms existing jailbreak techniques, revealing critical vulnerabilities in current defense mechanisms. Our findings underscore the necessity of a hybrid security strategy that integrates predefined security mechanisms with real-time semantic decomposition to enhance the security of LLMs.

📄 PDF Abstract BibTeX arXiv:2505.14316

Code (0)

등록된 구현이 없습니다.

Tasks

Question AnsweringText Generation

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음
Focus 설명 없음

Similar Papers 제목 키워드 기반

Bidirectional Intention Inference Enhances LLMs' Defense Against Multi-Turn Jailbreak Attacks

2025-09-25 · Haibo Tong, Dongcheng Zhao, Guobin Shen, Xiang He 외 arxiv

The remarkable capabilities of Large Language Models (LLMs) have raised significant safety concerns, particularly regarding "jailbreak" attacks that exploit adversarial prompts to bypass safety alignment mechanisms. Exis…

Breaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy Space

2025-05-27 · Yao Huang, Yitong Sun, Shouwei Ruan, Yichi Zhang 외

Large Language Models (LLMs), despite advanced general capabilities, still suffer from numerous safety risks, especially jailbreak attacks that bypass safety protocols. Understanding these vulnerabilities through black-b…

Prompt Engineering

Mitigating Jailbreaks with Intent-Aware LLMs

2025-08-16 · Wei Jie Yeo, Ranjan Satapathy, Erik Cambria arxiv

Despite extensive safety-tuning, large language models (LLMs) remain vulnerable to jailbreak attacks via adversarially crafted instructions, reflecting a persistent trade-off between safety and task performance. In this …

Hidden You Malicious Goal Into Benign Narratives: Jailbreak Large Language Models through Logic Chain Injection

2024-04-07 · Zhilong Wang, Yebo Cao, Peng Liu

Jailbreak attacks on Language Model Models (LLMs) entail crafting prompts aimed at exploiting the models to generate malicious content. Existing jailbreak attacks can successfully deceive the LLMs, however they cannot de…

Language ModelingLanguage Modelling

Understanding and Enhancing the Transferability of Jailbreaking Attacks

2025-02-05 · Runqi Lin, Bo Han, Fengwang Li, Tongling Liu

Jailbreaking attacks can effectively manipulate open-source large language models (LLMs) to produce harmful responses. However, these attacks exhibit limited transferability, failing to disrupt proprietary LLMs consisten…

Intent RecognitionRed Teaming