Huan Liu

This paper provides the first comprehensive, end-to-end survey dedicated to the security of Retrieval-Augmented Generation (RAG) systems, systematically mapping threats, defenses, and benchmarks across the entire pipeline.

The paper proposes MMGuard, a proactive defense mechanism that injects unlearnable, human-imperceptible perturbations into multimodal data to prevent unauthorized fine-tuning of Large Vision-Language Models (LVLMs).

While restricting a model to the theoretical Markov boundary can significantly improve prediction, the practical process of discovering and using this boundary is often computationally infeasible and does not consistently outperform using the full feature set.