ML Security PapersTowards Secure Retrieval-Augmented Generation: A Comprehensive Review of Threats, Defenses and Benchmarks
Yanming Mu 1,2, Hao Hu 1,2, Feiyang Li 1,2, Qiao Yuan 3, Jiang Wu 3, Zichuan Liu 3, Pengcheng Liu 3, Mei Wang 3, Hongwei Zhou 3, Yuling Liu 4
1 State Key Laboratory of Mathematical Engineering and Advanced Computing
2 Information Engineering University
Published on arXiv
2603.21654
Prompt Injection
OWASP LLM Top 10 — LLM01
Training Data Poisoning
OWASP LLM Top 10 — LLM03
Sensitive Information Disclosure
OWASP LLM Top 10 — LLM06
Key Finding
Provides unified analysis of RAG pipeline security threats and consolidates authoritative test datasets, security standards, and evaluation frameworks for future RAG security research
Retrieval-Augmented Generation (RAG) significantly mitigates the hallucinations and domain knowledge deficiency in large language models by incorporating external knowledge bases. However, the multi-module architecture of RAG introduces complex system-level security vulnerabilities. Guided by the RAG workflow, this paper analyzes the underlying vulnerability mechanisms and systematically categorizes core threat vectors such as data poisoning, adversarial attacks, and membership inference attacks. Based on this threat assessment, we construct a taxonomy of RAG defense technologies from a dual perspective encompassing both input and output stages. The input-side analysis reviews data protection mechanisms including dynamic access control, homomorphic encryption retrieval, and adversarial pre-filtering. The output-side examination summarizes advanced leakage prevention techniques such as federated learning isolation, differential privacy perturbation, and lightweight data sanitization. To establish a unified benchmark for future experimental design, we consolidate authoritative test datasets, security standards, and evaluation frameworks. To the best of our knowledge, this paper presents the first end-to-end survey dedicated to the security of RAG systems. Distinct from existing literature that isolates specific vulnerabilities, we systematically map the entire pipeline-providing a unified analysis of threat models, defense mechanisms, and evaluation benchmarks. By enabling deep insights into potential risks, this work seeks to foster the development of highly robust and trustworthy next-generation RAG systems.
Key Contributions
- First comprehensive end-to-end survey of RAG system security covering threats, defenses, and benchmarks
- Systematic taxonomy of RAG-specific vulnerabilities including data poisoning, adversarial attacks, and membership inference mapped to RAG workflow stages
- Dual-perspective defense framework analyzing input-side protections (access control, encryption, pre-filtering) and output-side techniques (federated learning, differential privacy, sanitization)