MAPLE-Guard：针对多智能体系统记忆链接投毒的记忆感知防御
原标题：MAPLE-Guard: Memory-Aware Link Enforcement Against Memory-Link Poisoning in Multi-Agent Systems
AI 摘要
MAPLE-Guard 是一种针对基于 LLM 的多智能体系统（MAS）中记忆链接投毒攻击的新型防御机制。它通过监控记忆生命周期，在写入、检索、提升和跨智能体重用等环节设置门控，以隔离风险记忆、过滤不安全检索并阻止中毒的私有记忆进入共享记忆。在 LongMemEval 和 AppWorld 基准测试中，MAPLE-Guard 将攻击成功率分别从 38.2% 降至 0.9% 和从 34.7% 降至 0.2%，同时显著提升了多智能体防御成功率。研究结果表明，记忆感知的链接执行能够弥补提示级和拓扑级防御的不足。
正文节选
Computer Science > Multiagent Systems Title:MAPLE-Guard: Memory-Aware Link Enforcement Against Memory-Link Poisoning in Multi-Agent Systems View PDF HTML (experimental) Abstract:LLM-based multi-agent systems (MAS) increasingly rely on persistent private and shared memories for long-horizon coordination. This memory layer improves continuity, but it also gives attackers a durable channel: a poisoned memory can be written once, continuously retrieved in later tasks, promoted into share