REVIEW 4 cited by
"Glue pizza and eat rocks" -- Exploiting Vulnerabilities in Retrieval-Augmented Generative Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Retrieval-Augmented Generative (RAG) models enhance Large Language Models (LLMs) by integrating external knowledge bases, improving their performance in applications like fact-checking and information searching. In this paper, we demonstrate a security threat where adversaries can exploit the openness of these knowledge bases by injecting deceptive content into the retrieval database, intentionally changing the model's behavior. This threat is critical as it mirrors real-world usage scenarios where RAG systems interact with publicly accessible knowledge bases, such as web scrapings and user-contributed data pools. To be more realistic, we target a realistic setting where the adversary has no knowledge of users' queries, knowledge base data, and the LLM parameters. We demonstrate that it is possible to exploit the model successfully through crafted content uploads with access to the retriever. Our findings emphasize an urgent need for security measures in the design and deployment of RAG systems to prevent potential manipulation and ensure the integrity of machine-generated content.
Forward citations
Cited by 4 Pith papers
-
Layer-Wise Perturbations via Sparse Autoencoders for Adversarial Text Generation
Sparse autoencoder activation perturbation (SFPF) applied on top of existing jailbreak prompts raises attack success rate on Qwen3-32B, but with no defense evaluation and weak reproducibility.
-
What do Large Language Models know about materials?
A benchmark-style study shows larger open LLMs reproduce element melting temperatures better, yet all tested models hallucinate values for elements lacking a melting point.
-
Benchmarking Poisoning Attacks against Retrieval-Augmented Generation
A unified benchmark evaluation finds that existing RAG poisoning attacks remain effective on standard QA datasets, drop on expanded knowledge bases, and are only partially mitigated by current defenses.
-
RAG Security and Privacy: Formalizing the Threat Model and Attack Surface
A formal RAG threat model is defined with four adversary classes and game-based notions of membership inference, leakage, and poisoning, but the definitions largely restate known concepts and the main DP-based protect...
Discussion (0). Continue with ORCID to comment.