MAS-Shield: A Defense Framework for Secure and Efficient LLM MAS

Large Language Model (LLM)-based Multi-Agent Systems (MAS) are susceptible to linguistic attacks that can trigger cascading failures across the network. Existing defenses face a fundamental dilemma: lightweight single-auditor methods are prone to single points of failure, while robust committee-based approaches incur prohibitive computational costs in multi-turn interactions. To address this challenge, we propose $\mathbf{\text{MAS-Shield}}$, a secure and efficient defense framework designed with a coarse-to-fine filtering pipeline. Rather than applying uniform scrutiny, MAS-Shield dynamically allocates defense resources through a three-stage protocol: (1) $\mathbf{\text{Critical}}\mathbf{\text{Agent}}\mathbf{\text{Selection}}$ strategically targets high-influence nodes to narrow the defense surface; (2) $\mathbf{\text{Light}}\mathbf{\text{Auditing}}$ employs lightweight sentry models to rapidly filter the majority of benign cases; and (3) $\mathbf{\text{Global}}\mathbf{\text{Consensus}}\mathbf{\text{Auditing}}$ escalates only suspicious or ambiguous signals to a heavyweight committee for definitive arbitration. This hierarchical design effectively optimizes the security-efficiency trade-off. Experiments demonstrate that MAS-Shield achieves a 92.5% recovery rate against diverse adversarial scenarios and reduces defense latency by over 70% compared to existing methods.