Metraipolis › archive › dc_bf244067dd13
DenialRAG: Single-Document RAG Poisoning via Embedded Parametric Denial — arXiv:2608.02678
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.02678
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:17:37.653Z
- SHA-256 of the archived text
3fb7c0696b9e09116dcc00010a71a6c5a1fd906f250f3c20537e971cb9862793- Stable URL of this page
https://metraipolis.com/documento/dc_bf244067dd13
How to cite
DenialRAG: Single-Document RAG Poisoning via Embedded Parametric Denial — arXiv:2608.02678 — arXiv (en), https://arxiv.org/abs/2608.02678 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bf244067dd13 (Metraipolis archive dc_bf244067dd13, sha256 3fb7c0696b9e09116dcc00010a71a6c5a1fd906f250f3c20537e971cb9862793)
BibTeX
@misc{dcbf244067dd13,
title = {DenialRAG: Single-Document RAG Poisoning via Embedded Parametric Denial — arXiv:2608.02678},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.02678},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bf244067dd13; sha256 3fb7c0696b9e09116dcc00010a71a6c5a1fd906f250f3c20537e971cb9862793},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.02678. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Retrieval-augmented generation (RAG) systems are vulnerable to corpus poisoning: an attacker who inserts a crafted document into the retrieval corpus can steer the underlying large language model (LLM) toward an attacker-chosen wrong answer. Prior single-document attacks typically avoid explicitly naming and refuting the correct answer inside the poisoned passage. In this paper, we examine a complementary design and propose \emph{DenialRAG}, a single-document poisoning attack that explicitly names the correct answer, denies it, and presents an attacker-controlled explanation for favoring the wrong answer. By placing both the correct answer and the corresponding poisoned answer inside the same retrieved passage, DenialRAG embeds the conflict directly into the context seen by the generator. We evaluate DenialRAG against four published single-document poisoning attacks across three open-domain question-answering datasets, eight target LLMs from four vendors, and five inference-time defenses. The results show that attack effectiveness is strongly model-dependent: DenialRAG achieves the highest attack success rate (ASR) on all three Mistral-7B datasets and remains effective on several other target LLMs, while other attacks dominate in some model regimes. Defense results show meaningful ASR reductions but non-uniform protection, with each defense leaving residual ASR in some settings. Component-level and cross-model analyses further identify the embedded denial as the most influential tested component and show that different poisoning mechanisms lose effectiveness at different rates across model groups. Together, these results show that RAG poisoning risk cannot be fully characterized by a single attack family or a single target model.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.AI).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.02678 — 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 2026-10-31T12:05:52.927Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).