paper-with-me

홈 › Papers

Can LLMs Deeply Detect Complex Malicious Queries? A Framework for Jailbreaking via Obfuscating Intent

2024-05-06 · Shang Shang, Xinqiang Zhao, Zhongjiang Yao, Yepeng Yao, Liya Su, Zijing Fan, Xiaodan Zhang, Zhengwei Jiang

To demonstrate and address the underlying maliciousness, we propose a theoretical hypothesis and analytical approach, and introduce a new black-box jailbreak attack methodology named IntentObfuscator, exploiting this identified flaw by obfuscating the true intentions behind user prompts.This approach compels LLMs to inadvertently generate restricted content, bypassing their built-in content security measures. We detail two implementations under this framework: "Obscure Intention" and "Create Ambiguity", which manipulate query complexity and ambiguity to evade malicious intent detection effectively. We empirically validate the effectiveness of the IntentObfuscator method across several models, including ChatGPT-3.5, ChatGPT-4, Qwen and Baichuan, achieving an average jailbreak success rate of 69.21\%. Notably, our tests on ChatGPT-3.5, which claims 100 million weekly active users, achieved a remarkable success rate of 83.65\%. We also extend our validation to diverse types of sensitive content like graphic violence, racism, sexism, political sensitivity, cybersecurity threats, and criminal skills, further proving the substantial impact of our findings on enhancing 'Red Team' strategies against LLM content security frameworks.

📄 PDF Abstract BibTeX arXiv:2405.03654

Code (0)

등록된 구현이 없습니다.

Tasks

Intent Detection

Similar Papers 제목 키워드 기반

Adversarial Distilled Retrieval-Augmented Guarding Model for Online Malicious Intent Detection

2025-09-18 · Yihao Guo, Haocheng Bian, Liutong Zhou, Ze Wang 외 arxiv

With the deployment of Large Language Models (LLMs) in interactive applications, online malicious intent detection has become increasingly critical. However, existing approaches fall short of handling diverse and complex…

Out-of-Distribution DetectionIntent Detection

Jailbreaking for the Average Jane: Choosing Optimal Jailbreaks via Bandit Algorithms for Automatically Enhanced Queries

2026-06-25 · Prarabdh Shukla, Ritik, Suhas Rao, Arpit Agarwal 외 arxiv

With a profusion of jailbreaks for LLMs now widely known, a growing concern is that non-expert malicious actors ("the average Jane") could elicit actionable responses to malicious requests. In this work, we examine wheth…

DrAttack: Prompt Decomposition and Reconstruction Makes Powerful LLM Jailbreakers

2024-02-25 · Xirui Li, Ruochen Wang, Minhao Cheng, Tianyi Zhou 외

The safety alignment of Large Language Models (LLMs) is vulnerable to both manual and automated jailbreak attacks, which adversarially trigger LLMs to output harmful content. However, current methods for jailbreaking LLM…

In-Context LearningSafety Alignment

Enhancing GraphQL Security by Detecting Malicious Queries Using Large Language Models, Sentence Transformers, and Convolutional Neural Networks

2025-08-14 · Irash Perera, Hiranya Abeyrathne, Sanjeewa Malalgoda, Arshardh Ifthikar arxiv

GraphQL's flexibility, while beneficial for efficient data fetching, introduces unique security vulnerabilities that traditional API security mechanisms often fail to address. Malicious GraphQL queries can exploit the la…

InfoFlood: Jailbreaking Large Language Models with Information Overload

2025-06-13 · Advait Yadav, Haibo Jin, Man Luo, Jun Zhuang 외

Large Language Models (LLMs) have demonstrated remarkable capabilities across various domains. However, their potential to generate harmful responses has raised significant societal and regulatory concerns, especially wh…