テーマ: poisoning
-
Poisoned retrieval corpora: how a few planted documents can steer a RAG system's answers
Retrieval-augmented generation trusts whatever the retriever returns. Research has shown that injecting a small number of crafted texts into a knowledge base can make a system give an attacker-chosen answer to a targeted question. Defences are about who can write to the corpus, provenance per passage and answer checks.
機械可読: JSON