揭示RAG系统三大攻击风险并提出防护清单
Adversarial Threat Vectors and Risk Mitigation for Retrieval-Augmented Generation Systems
- 从提示注入、数据污染等角度分析RAG安全威胁
- 提出包含输入验证等三类控制措施的优先级清单
- 适合关注大模型应用安全的工程师与研究者
检索增强生成（RAG）系统通过整合大型语言模型（LLMs）与外部知识源，在工业界广泛应用。然而，这类系统面临多种对抗性攻击风险。本文基于近期行业采用趋势，识别出三大主要攻击向量：提示注入、数据污染和对抗性查询操纵。我们从风险管理视角分析这些威胁，并提出一套优先级明确的鲁棒控制清单，包括输入验证、对抗训练和实时监控等风险缓解措施。
原文摘要 · Abstract (English)
Retrieval-Augmented Generation (RAG) systems, which integrate Large Language Models (LLMs) with external knowledge sources, are vulnerable to a range of adversarial attack vectors. This paper examines the importance of RAG systems through recent industry adoption trends and identifies the prominent attack vectors for RAG: prompt injection, data poisoning, and adversarial query manipulation. We analyze these threats under risk management lens, and propose robust prioritized control list that includes risk-mitigating actions like input validation, adversarial training, and real-time monitoring.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。