arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2505.15420cs.CRcs.AI
收藏

Silent Leaks:通过良性查询对RAG系统进行的隐式知识提取攻击

Silent Leaks: Implicit Knowledge Extraction Attack on RAG Systems through Benign Queries

National University of Singapore(新加坡国立大学);Peking University(北京大学);Tsinghua University(清华大学)

Yuhao Wang, Wenjie Qu, Shengfang Zhai, Yanze Jiang, Zichen Liu, Yue Liu, Yinpeng Dong, Jiaheng Zhang

AI总结:

本文提出针对RAG系统的隐式知识提取攻击IKEA,通过锚定概念生成良性查询,结合经验反思采样与信任域定向变异机制探索知识库,提取效率和成功率远超基线,可构建性能接近原系统的替代RAG,暴露了RAG系统的隐蔽版权风险。

摘要:

检索增强生成(RAG)系统通过整合外部知识库来增强大语言模型(LLM),但这可能使其面临提取攻击,引发潜在的版权与隐私风险。然而,现有提取方法通常依赖提示注入、越狱等恶意输入,很容易通过输入级或输出级检测被发现。本文提出隐式知识提取攻击(Implicit Knowledge Extraction Attack,IKEA),通过良性查询对RAG系统实施知识提取。具体而言,IKEA首先利用锚定概念——即与内部知识相关的关键词——生成外观自然的查询,随后设计两种机制引导锚定概念充分“探索”RAG的知识库:(1)经验反思采样,基于过往查询-响应历史对锚定概念进行采样,确保其与主题的相关性;(2)信任域定向变异,在相似性约束下对锚定概念进行迭代变异,进一步挖掘嵌入空间。大量实验证明了IKEA在多种防御手段下的有效性,其提取效率超越基线方法80%以上,攻击成功率超越基线90%以上。此外,利用IKEA提取内容构建的替代RAG系统性能与原RAG相当,在多项评估任务中表现优于基于基线方法构建的替代系统,凸显了RAG系统中隐蔽的版权侵权风险。

英文摘要:

Retrieval-Augmented Generation (RAG) systems enhance large language models (LLMs) by incorporating external knowledge bases, but this may expose them to extraction attacks, leading to potential copyright and privacy risks. However, existing extraction methods typically rely on malicious inputs such as prompt injection or jailbreaking, making them easily detectable via input- or output-level detection. In this paper, we introduce Implicit Knowledge Extraction Attack (IKEA), which conducts Knowledge Extraction on RAG systems through benign queries. Specifically, IKEA first leverages anchor concepts-keywords related to internal knowledge-to generate queries with a natural appearance, and then designs two mechanisms that lead anchor concepts to thoroughly "explore" the RAG's knowledge: (1) Experience Reflection Sampling, which samples anchor concepts based on past query-response histories, ensuring their relevance to the topic; (2) Trust Region Directed Mutation, which iteratively mutates anchor concepts under similarity constraints to further exploit the embedding space. Extensive experiments demonstrate IKEA's effectiveness under various defenses, surpassing baselines by over 80% in extraction efficiency and 90% in attack success rate. Moreover, the substitute RAG system built from IKEA's extractions shows comparable performance to the original RAG and outperforms those based on baselines across multiple evaluation tasks, underscoring the stealthy copyright infringement risk in RAG systems.

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。

↑