Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.08304 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914608261890048 |
|---|---|
| author | Xu, Yuming Zhang, Mingtao Ge, Zhuohan Li, Haoyang Hu, Nicole Zhang, Yongqi Wen, Zhiyuan Zhang, Jason Chen Li, Qing Chen, Lei |
| author_facet | Xu, Yuming Zhang, Mingtao Ge, Zhuohan Li, Haoyang Hu, Nicole Zhang, Yongqi Wen, Zhiyuan Zhang, Jason Chen Li, Qing Chen, Lei |
| contents | Retrieval-augmented generation (RAG) extends large language models (LLMs) with external knowledge, but this access path also introduces security risks that existing work often conflates with inherent LLM flaws. We frame secure RAG as securing external knowledge access and organize the literature with SLOT, a taxonomy along four axes: the attack Surface (S) where an adversary acts, the defense Layer (L) that controls the same point, the Objective (O) it breaks following the CIA properties, and the Target (T) it pursues, from a single known query (T1) to target-claim manipulation across a query distribution (T2). Mapping attacks, defenses, remediation, and evaluation onto a six-stage knowledge-access pipeline, we expose two structural mismatches. Finally, we discuss directions for more realistic targets, no-blind-spot and adaptively evaluated defenses, stronger confidentiality, and evaluation for multimodal and agentic RAG. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_08304 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Securing Retrieval-Augmented Generation: A Taxonomy of Attacks, Defenses, and Future Directions Xu, Yuming Zhang, Mingtao Ge, Zhuohan Li, Haoyang Hu, Nicole Zhang, Yongqi Wen, Zhiyuan Zhang, Jason Chen Li, Qing Chen, Lei Cryptography and Security Artificial Intelligence Retrieval-augmented generation (RAG) extends large language models (LLMs) with external knowledge, but this access path also introduces security risks that existing work often conflates with inherent LLM flaws. We frame secure RAG as securing external knowledge access and organize the literature with SLOT, a taxonomy along four axes: the attack Surface (S) where an adversary acts, the defense Layer (L) that controls the same point, the Objective (O) it breaks following the CIA properties, and the Target (T) it pursues, from a single known query (T1) to target-claim manipulation across a query distribution (T2). Mapping attacks, defenses, remediation, and evaluation onto a six-stage knowledge-access pipeline, we expose two structural mismatches. Finally, we discuss directions for more realistic targets, no-blind-spot and adaptively evaluated defenses, stronger confidentiality, and evaluation for multimodal and agentic RAG. |
| title | Securing Retrieval-Augmented Generation: A Taxonomy of Attacks, Defenses, and Future Directions |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2604.08304 |