Evaluated against the non-adaptive attacker described in the original PoisonedRAG paper, the full pipeline reduces attack success rate from roughly 91% to roughly 13%, while preserving accuracy on benign, unpoisoned queries.
Abstract
Retrieval-Augmented Generation (RAG) lets a large language model answer questions using documents retrieved from an external knowledge base at query time. This makes RAG useful for private data, fast-changing information, and reducing hallucination, but it also means the model's answer is only as trustworthy as whatever the retriever hands it. If the knowledge base accepts writes from more than one party, an attacker needs only a handful of adversarial documents to steer the model toward a chosen wrong answer. PoisonedRAG demonstrated this: as few as five crafted documents flip an undefended system's answer roughly 90% of the time, and three natural single-stage defenses (perplexity filtering, query paraphrasing, knowledge-base expansion) leave attack success at 30% or higher. We built TriShieldRAG to close that gap. Rather than relying on one checkpoint, we place three independent, formally specified rings across the pipeline: an Ingest Guard that screens documents for lexical and statistical poisoning signatures; a Retrieval Scorer that re-ranks the retrieved set by a provenance and consistency-weighted trust score; and a Cross-LLM Consensus stage that polls three architecturally diverse language models (Claude, Mistral Small, Llama 3.2) and allows one bounded re-retrieval on disagreement. We derive the conditions under which Rings 2 and 3 are expected to work: a minority-poison assumption and an explicit provenance-tag assumption. Our reported configuration is consistent with this analysis, though we have not yet run the controlled poison-fraction sweep needed to confirm it independently. Evaluated against the non-adaptive attacker from the original PoisonedRAG, over a 5,000-document Wikipedia knowledge base with 10 target questions, the full pipeline reduces attack success rate from roughly 91% to roughly 13% while preserving accuracy on benign queries.
Retrieval-augmented generation (RAG) improves the factuality of large language models by grounding responses in external documents, but it also exposes a critical security vulnerability: adversarial documents injected into the knowledge database can enter the context window and steer the model toward targeted incorrect answers. Existing post-retrieval defenses rely on instruction following, parametric knowledge, or text-level consistency, all of which can be imitated or optimized against by adaptive attackers. We propose RAGSentinel, a training-free, label-free defense for black-box RAG systems. RAGSentinel uses a surrogate encoder to measure query-conditioned hidden-state shifts induced by retrieved documents, removes shared topic directions, and filters poisoned documents as geometric outliers from a robust majority consensus. We prove that, under an honest-majority assumption and a representation-level separation condition, RAGSentinel exactly recovers a poison-free majority-sized context. Experiments across three question-answering datasets, three LLM families, and multiple poisoning attacks show that RAGSentinel consistently achieves low attack success rates while preserving competitive accuracy and remaining effective against adaptive attacks with full pipeline knowledge.
Yueyang Quan, Anjun Gao, Yu Xia et al.· 0 citations
Retrieval-Augmented Generation (RAG) grounds LLM answers in query-time retrieved documents, so reliability depends on what the retriever returns. PoisonedRAG (Zou et al., USENIX Security'25) showed five crafted documents mislead an undefended system in nearly 90% of cases, and that single-stage defenses give limited robustness. We propose TriShieldRAG, a three-layered framework: an Ingest Guard for document-level screening, a Retrieval Scorer for trust-aware re-ranking, and a Cross-LLM Consensus over three diverse models. We reasoned that collectively screening, re-ranking and validating retrieved evidence would give complementary protection, limiting the ability of poisoned documents to succeed through any single failure. We evaluate against non-adaptive and adaptive poisoning. Non-adaptively, on the full 2.68M-passage Natural Questions (NQ) corpus with the original PoisonedRAG attack, it cuts attack success from 79 +/- 1.0% to 1 +/- 0.0%. Adaptive attacks expose fundamental limits of layering. By changing only the document formatting, without modifying the poison text or accessing the retriever, the attacker reduces the Ingest Guard score from 0.500 to 0.000 and bypasses it on all 500 tested documents across three corpora. The remaining layers then give no protection: 62 +/- 0.8% attack success against a 56 +/- 2.5% undefended baseline on NQ, and 85 +/- 0.6% against 86 +/- 0.6% on HotpotQA. Layered defenses relying on the same retrieved evidence fail together: poisoned context misleads both re-ranking and consensus validation. Minority-poison thresholds prove corpus-dependent, at 0.214, 0.251 and 0.558 rather than the derived 0.5; a closed form we proposed for these failed a pre-registered prediction and is retracted. Cross-model agreement is misleading, reaching 0.96 while attack success approaches 99%. We release the framework, the evasion-certification methodology and artifacts.
S. K. Mohanty, Rohit Patel, K. Yuvaraj et al.· 0 citations
Retrieval-Augmented Generation (RAG) systems typically consist of a dense retrieval (DR) model for initial retrieval and a neural ranking model (NRM) for re-ranking.Existing robustness studies in RAG mainly focus on NRMs, while adversarial attacks on DR models are mostly limited to word-level perturbations.For low-ranked target documents that are irrelevant to the query, simple word-level attacks are insufficient to mislead DR models into substantially promoting their rankings.To solve these problems, we propose SentAttack, a sentence-level black-box adversarial attack method for DR models.SentAttack is designed as a two-stage method.In the first stage, SentAttack interacts with the black-box RAG system via iterative retrieval to collect ranked documents and ranking information for training a surrogate DR model.In the second stage, SentAttack uses the surrogate DR model to encode and cluster documents relevant to the target query, yielding multiple cluster centroids.These centroids are concatenated with the target document at the sentence level to form an initial set of adversarial candidates.SentAttack then optimizes these candidates using a query- and centroid-guided objective combined with gradient-guided beam search.Extensive experiments demonstrate that SentAttack outperforms existing adversarial attacks on DR models, with especially strong performance on low-ranked target documents.
Luping Wei, Yamin Hu, Sihan Shang et al.· 0 citations
RAGuard, a layered defense against corpus-poisoning attacks on RAG pipelines, is introduced, showing that keyword-preserving poisons leave lexical retrievers such as BM25 essentially unaffected, an observation that delineates the boundary of the threat model.
Retrieval-augmented generation (RAG) answers a question by retrieving passages from a vector store and trusting them as context, so anyone who can add documents can try to steer the answer. A recent, appealing defense filters poisoning at ingestion, rejecting any document that behaves like a hub. We show it -- and every ingestion-time filter -- is defeated by a coordinated adversary that injects a handful of individually unremarkable documents which together surround one target query and seize its top-k (on BGE-large / BEIR, m=10 documents take 10/10; 9.9/10 on a live HNSW index). The attack is not theoretical. Realized as ordinary fluent text and run end-to-end through a BGE-large + HNSW + Qwen2.5-7B pipeline, it makes the generator emit the attacker's planted claim in 88% of targets, versus 0% without the injection. And no admission-time defense stops it: at ingestion an attack cone is geometrically identical to a legitimate niche upload, so -- measuring this directly -- the strongest trained classifier, given every feature and thousands of examples, separates the two no better than chance, catching 4.2% of attacks at a 1% false-positive rate. We prove this limit for the entire class of ingestion-time statistics (any decision from documents and reference queries alone), and it reproduces -- and worsens -- across two corpora and five encoders. The one signal that separates an attack from legitimate niche ingestion -- a query's demand -- is invisible before retrieval, which is also the escape: a retrieval-time detector that observes demand catches 100% of the attacks at the same 1% false-positive rate. Coverage of the query space by an admission gate is not containment of coordinated poisoning; robust defense must move past the front door, to demand.
Multimodal Retrieval Augmented Generation (M-RAG) is increasingly vulnerable to adversarial attacks where malicious data are crafted to produce embeddings that align with benign entries in the vector space, deceiving retrieval and inducing harmful outputs. Existing defenses primarily operate at query time, relying on auxiliary detectors, similarity re-ranking, or feature-consistency checks. However, these approaches suffer from non-trivial inference overhead, generalize poorly to unseen attack strategies, and often assume specific attack distributions. To address this, we propose DSPrompt, a Dynamic Soft Prompt defense framework that directly reshapes the retriever's embedding semantics, without modifying the retrieval pipeline. It inserts few learnable soft prompts into each layer of the visual and textual encoders of a frozen retriever, utilizing a shallow-to-deep length schedule that is adaptive to the capacity in the model layers. These prompts are trained under a dynamic min-max scheme: an online multimodal attacker continually crafts hard adversarial documents against the current retriever, while the defender is updated to push such documents out of the top-k while preserving the ranking and diversity of benign evidence. Because the defended encoder can be pre-computed and indexed exactly as in standard dense retrieval, DSPrompt incurs no additional per-query optimization and introduces fewer than 1% additional parameters. Extensive experiments across four benchmarks and three representative poisoning attacks show that DSPrompt substantially reduces the attack success rate and poison retrieval rate while maintaining near-lossless retrieval utility and generation fidelity, consistently outperforming existing defense baselines at a fraction of their computational cost.
Chang Liu, Y. Lai, Mingyue Cui et al.· 0 citations