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 2605.28645cs.CRcs.CL
收藏

GraphSteal:通过遍历重建从图RAG中窃取结构知识

GraphSteal: Structural Knowledge Stealing from Graph RAG via Traversal Reconstruction

School of Computer Science, Shanghai Jiao Tong University(上海交通大学计算机科学学院)

Jinze Gu, Qinghua Mao, Xi Lin, Jun Wu

AI总结:

提出一种结构导向的重建框架,通过深度优先启发式搜索和广度优先扩散搜索,从黑盒图RAG系统中高保真恢复隐藏知识图,揭示敏感实体、关系和结构依赖。

摘要:

检索增强生成(RAG)通过将生成过程基于查询相关的外部证据来增强LLM。除了非结构化文本语料库外,图RAG将知识图谱集成到检索流程中,使LLM能够访问编码在结构化知识中的实体、关系和多跳依赖。然而,赋能图RAG的相同结构化知识也创造了新的隐私攻击面。我们证明,图RAG系统可以转变为结构预言机:通过自适应黑盒交互,对手可以引出足够的关联证据,以重建隐藏知识图的实质性部分。我们提出了一种面向结构的重建框架,从局部和全局角度恢复目标图。具体来说,深度优先启发式搜索通过递归扩展以实体为中心的证据来提取细粒度节点属性,而广度优先扩散搜索通过跨关系诱导邻域传播来推断图拓扑。在通用和医疗场景上的实验表明,我们的方法可以从代表性图RAG系统中恢复超过90%的原始知识图,以高保真度揭示敏感实体、关系和结构依赖。现有的防护措施对我们的攻击提供的防御有限,突显了在图RAG流程中保护结构隐私的固有困难。

英文摘要:

Retrieval-Augmented Generation (RAG) enhances LLMs by grounding generation in query-relevant external evidence. Beyond unstructured text corpora, Graph RAG integrates knowledge graphs into the retrieval pipeline, enabling LLMs to access entities, relations, and multi-hop dependencies encoded in structured knowledge. However, the same structured knowledge that empowers Graph RAG also creates a new privacy attack surface. We demonstrate that Graph RAG systems can be turned into structural oracles: through adaptive black-box interactions, an adversary can elicit sufficient relational evidence to reconstruct substantial portions of the hidden knowledge graph. We propose a structure-oriented reconstruction framework that recovers targeted graphs from both local and global perspectives. Specifically, Depth-Wise Heuristic Search extracts fine-grained node attributes by recursively expanding entity-centered evidence, while Breadth-Wise Diffusion Search infers graph topology by propagating across relation-induced neighborhoods. Experiments on generic and healthcare scenarios demonstrate that our method can recover over 90\% of the original knowledge graph from representative Graph RAG systems, revealing sensitive entities, relations, and structural dependencies with high fidelity. Existing guradrails provide limited defense against our attack, highlighting the inherent difficulty of safeguarding structural privacy in Graph RAG pipelines.

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。

↑