Five Queries Are Enough: Query-Efficient and Surrogate-Free Membership Inference Attacks on RAG via Entailment

Nguyen Linh Bao Nguyen, Wanlun Ma, and Viet Vo, Swinburne University of Technology, Australia; Alsharif Abuadbba, CSIRO, Australia; Minghong Fang, University of Louisville, USA; Jun Zhang and Yang Xiang, Swinburne University of Technology, Australia

Retrieval‑Augmented Generation (RAG) has become central to large language model (LLM) deployments, grounding responses in enterprise or proprietary data to reduce hallucinations. However, this design introduces a new privacy risk: model outputs may signal the presence of specific documents in the retrieval corpus, enabling membership inference attacks (MIAs) that leak sensitive information. Existing MIAs are feasible, but they often rely on easily-detected templated queries or require many non-templated yet costly and repetitive queries, limiting practicality. We ask: Can an adversary launch a limited-budget, surrogate-free, stealthy, and defense-agnostic membership inference attack using non-templated queries? We present MEntA (Membership Entailment Attack)—a query‑efficient MIA that leverages natural‑language entailment to maximize information gained per query. By asking low-cost, broad, information‑seeking questions and measuring entailment between model responses and candidate documents, MEntA eliminates the need for costly shadow LLMs and large query budgets. Across NFCorpus, SCIDOCS, and TREC‑COVID, MEntA achieves up to 0.991 AUC with only 5 queries, outperforming prior methods by up to 0.42 AUC under equivalent conditions. It remains effective under state‑of‑the‑art (SOTA) RAG defenses, while current detectors miss MEntA or flag benign queries at high rates. Regarding cost, MEntA reduces total attack cost by up to 65× lower compared to SOTA attacks under the same attack setting. Our findings expose the feasibility of realistic, low‑cost privacy leakage in RAG systems and highlight the urgent need for privacy‑aware retrieval and defense mechanisms.

USENIX Security '26 Open Access Sponsored by
King Abdullah University of Science and Technology (KAUST)

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

BibTeX
@inproceedings {320671,
author = {Nguyen Linh Bao Nguyen and Wanlun Ma and Viet Vo and Alsharif Abuadbba and Minghong Fang and Jun Zhang and Yang Xiang},
title = {Five Queries Are Enough: {Query-Efficient} and {Surrogate-Free} Membership Inference Attacks on {RAG} via Entailment},
booktitle = {35th USENIX Security Symposium (USENIX Security 26)},
year = {2026},
isbn = {978-1-939133-58-8},
address = {Baltimore, MD},
pages = {1147--1166},
url = {https://www.usenix.org/conference/usenixsecurity26/presentation/nguyen-nguyen},
publisher = {USENIX Association},
month = aug
}

Presentation Video