DenialRAG: Single-Document RAG Poisoning via Embedded Parametric Denial
Quick summary
arXiv:2608.02678v1 Announce Type: cross Abstract: 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
Key takeaways
- arXiv:2608.02678v1 Announce Type: cross Abstract: 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
Why it matters
“DenialRAG: Single-Document RAG Poisoning via Embedded Parametric Denial” highlights the need for repeatable measurement rather than a single impressive demonstration. Independent validation across datasets and clearly stated limitations determine whether a result can guide product decisions.
Member comments