Metraipolis › archive › dc_eba7fddf1e82
RAG-CT: Mitigating Privacy Risks on Retrieval-Augmented Generation Systems via Scanning Prompt Distribution — arXiv:2609.16095
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.16095
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T18:32:39.604Z
- SHA-256 of the archived text
d3fc32a7abdd3834b02dc4701534dbc26fccd7f0d4b274961554bcc30f81fdcc- Stable URL of this page
https://metraipolis.com/documento/dc_eba7fddf1e82
How to cite
RAG-CT: Mitigating Privacy Risks on Retrieval-Augmented Generation Systems via Scanning Prompt Distribution — arXiv:2609.16095 — arXiv (en), https://arxiv.org/abs/2609.16095 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_eba7fddf1e82 (Metraipolis archive dc_eba7fddf1e82, sha256 d3fc32a7abdd3834b02dc4701534dbc26fccd7f0d4b274961554bcc30f81fdcc)
BibTeX
@misc{dceba7fddf1e82,
title = {RAG-CT: Mitigating Privacy Risks on Retrieval-Augmented Generation Systems via Scanning Prompt Distribution — arXiv:2609.16095},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.16095},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_eba7fddf1e82; sha256 d3fc32a7abdd3834b02dc4701534dbc26fccd7f0d4b274961554bcc30f81fdcc},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.16095. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for improving the quality of generated contents of Large Language Models (LLMs) by grounding responses in external knowledge, thus reducing hallucinations and factual errors. However, recent studies have highlighted a critical vulnerability: adversaries can exploit the retrieval process to extract personally identifiable information (PII) from the underlying corpus. To mitigate this risk, we propose a novel defense, RAG-CT, that identifies malicious queries by analyzing their entropy and margin distributions and using a score-based detection method. Extensive experiments with four state-of-the-art attack strategies and four defense baselines on two datasets show that our approach significantly reduces PII leakage while outperforming existing defenses. This work provides a lightweight yet effective mechanism to protect RAG systems against PII leakage without requiring modifications to the underlying LLM or retriever. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-16 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR). Pagina del lavoro: https://arxiv.org/abs/2609.16095 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-15T18:03:09.586Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).