Vis-Poison: Poisoning Visual Knowledge in Multimodal Retrieval-Augmented Generation
Quick summary
arXiv:2608.20756v1 Announce Type: cross Abstract: While multimodal retrieval-augmented generation (RAG) systems increasingly rely on images as external knowledge sources, the introduction of poisoned visual evidence can severely compromise multimodal large language model (MLLM) generation. Unlike prior attacks that rely on altering textual metadata, we introduce Vis-Poison, a novel visual knowledge poisoning attack where the poisoned image itself is the attacker-controlled payload, without manipulating captions, summaries, metadata, or other associated text. Specifically, this attack is instan
Key takeaways
- arXiv:2608.20756v1 Announce Type: cross Abstract: While multimodal retrieval-augmented generation (RAG) systems increasingly rely on images as external knowledge sources, the introduction of poisoned visual evidence can severely compromise multimodal large language model (MLLM) generation.
- Unlike prior attacks that rely on altering textual metadata, we introduce Vis-Poison, a novel visual knowledge poisoning attack where the poisoned image itself is the attacker-controlled payload, without manipulating captions, summaries, metadata, or other associated text.
- Specifically, this attack is instan
Why it matters
This model development creates a new option for users and a new testing obligation for developers. A fixed evaluation set comparing quality, cost and failure behavior is more useful than launch claims.
Member comments