Newest papers

RAGLeak: Supply-Chain Privacy Backdoors in Retrieval-Augmented Generation

Yuchen Chen, Wei Cheng, Yuan Xiao, Junda He, Jieke Shi, 欧阳萍萍, Quanjun Zhang, Chunrong Fang, David Lo, Zhenyu Chen · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.25937 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

A geração aumentada de recuperação (RAG) permite que modelos de idiomas grandes acessem conhecimento externo durante a inferência, mas sua dependência de bases de conhecimento externas também introduz novas superfícies de ataque. Os ataques de extração de dados existentes contra sistemas RAG dependem de injeção rápida, que depende fortemente da capacidade de seguimento de instruções do gerador, ou backdoors que reproduzem ou parafraseiam documentos recuperados, mas lutam para equilibrar a fidelidade de extração e a stealthiness. Neste artigo, propomos RAGLeak, um ataque de backdoor de privacidade da cadeia de suprimentos que desdobla a fuga de informações da exposição direta ao texto. Quando desencadeado, o gerador obfusca reversivelmente documentos privados recuperados para recuperação de alta fidelidade pelo auxiliar LLM

Machine translation (Meta M2M100); the English original follows.

Abstract

Retrieval-Augmented Generation (RAG) enables large language models to access external knowledge during inference, but its reliance on external knowledge bases also introduces new attack surfaces. Existing data extraction attacks against RAG systems rely on either prompt injection, which heavily depends on the generator's instruction-following capability, or backdoors that reproduce or paraphrase retrieved documents but struggle to balance extraction fidelity and stealthiness. In this paper, we propose RAGLeak, a supply-chain privacy backdoor attack that decouples information leakage from direct textual exposure. When triggered, the generator reversibly obfuscates retrieved private documents for high-fidelity recovery by an auxiliary LLM. Across three model families and three application domains, RAGLeak achieves up to $100\%$ ASR with limited impact on benign utility and at most a $10$-percentage-point ASR drop under representative detection-based defenses. The backdoor remains effective after clean fine-tuning and across model scales and versions, revealing an underexplored supply-chain privacy risk in RAG systems. The source code is available at https://anonymous.4open.science/r/RAGLeak-4B67.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Backdoor, Computer science, Computer security, Paraphrase, Generator (circuit theory), Source code

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange