PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Models

Name: PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Models
Duration: 15 min
Description: Talk from 34th USENIX Security Symposium (USENIX Security '25).

Wei Zou

34th USENIX Security Symposium (USENIX Security '25) · Day 2 · LLM Security 3