arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2312.10982cs.CR
收藏

大型语言模型中攻击技术、实现与缓解策略的综合综述

A Comprehensive Survey of Attack Techniques, Implementation, and Mitigation Strategies in Large Language Models

Aysan Esmradi, Daniel Wankit Yip, Chun Fai Chan

AI总结:

本文综述 100 余篇研究,系统梳理针对 LLM 本身及其应用的攻击技术、真实实现、缓解措施与未来防御方向。

摘要:

尽管大型语言模型(LLM)广受欢迎,确保其安全性仍是一项持续挑战。开发者致力于增强 LLM 安全性,但漏洞依然存在,即使在 GPT-4 等先进版本中也是如此。攻击者利用这些弱点,凸显出在 AI 模型开发中采取主动网络安全措施的必要性。本文探讨两类攻击:针对模型本身的攻击和针对模型应用的攻击。前者需要专业知识、对模型数据的访问权限以及大量实现时间,而后者对攻击者来说更易实施,并受到越来越多关注。我们的研究综述了 100 多篇近期研究成果,对每种攻击类型进行深入分析。我们识别最新攻击方法,并探讨实施这些攻击的各种途径。我们全面研究缓解技术,评估其有效性与局限性。此外,我们总结了针对这些攻击的未来防御措施。我们还考察了真实世界技术,包括已报告的以及我们自行实现的针对 LLM 的攻击,以整合研究发现。我们的研究强调解决安全问题的紧迫性,旨在增进对 LLM 攻击的理解,并为这一不断发展领域中的稳健防御开发作出贡献。

英文摘要:

Ensuring the security of large language models (LLMs) is an ongoing challenge despite their widespread popularity. Developers work to enhance LLMs security, but vulnerabilities persist, even in advanced versions like GPT-4. Attackers exploit these weaknesses, highlighting the need for proactive cybersecurity measures in AI model development. This article explores two attack categories: attacks on models themselves and attacks on model applications. The former requires expertise, access to model data, and significant implementation time, while the latter is more accessible to attackers and has seen increased attention. Our study reviews over 100 recent research works, providing an in-depth analysis of each attack type. We identify the latest attack methods and explore various approaches to carry them out. We thoroughly investigate mitigation techniques, assessing their effectiveness and limitations. Furthermore, we summarize future defenses against these attacks. We also examine real-world techniques, including reported and our implemented attacks on LLMs, to consolidate our findings. Our research highlights the urgency of addressing security concerns and aims to enhance the understanding of LLM attacks, contributing to robust defense development in this evolving domain.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑