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

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Kaixiang, Zhou, Zhaojiacheng, Suvonov, Bunyod, Lou, Jiong, LI, Jie
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912868010557440
author Wang, Kaixiang
Zhou, Zhaojiacheng
Suvonov, Bunyod
Lou, Jiong
LI, Jie
author_facet Wang, Kaixiang
Zhou, Zhaojiacheng
Suvonov, Bunyod
Lou, Jiong
LI, Jie
contents 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 \textbf{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) \textbf{Critical Agent Selection } strategically targets high-influence nodes to narrow the defense surface; (2) \textbf{Light Auditing} employs lightweight sentry models to rapidly filter the majority of benign cases; and (3) \textbf{Global Consensus 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.
format Preprint
id arxiv_https___arxiv_org_abs_2511_22924
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MAS-Shield: A Defense Framework for Secure and Efficient LLM MAS
Wang, Kaixiang
Zhou, Zhaojiacheng
Suvonov, Bunyod
Lou, Jiong
LI, Jie
Multiagent Systems
Artificial Intelligence
Cryptography and Security
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 \textbf{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) \textbf{Critical Agent Selection } strategically targets high-influence nodes to narrow the defense surface; (2) \textbf{Light Auditing} employs lightweight sentry models to rapidly filter the majority of benign cases; and (3) \textbf{Global Consensus 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.
title MAS-Shield: A Defense Framework for Secure and Efficient LLM MAS
topic Multiagent Systems
Artificial Intelligence
Cryptography and Security
url https://arxiv.org/abs/2511.22924