基于大型语言模型 (LLM) 的多代理系统越来越依赖中间通信来协调复杂的任务。虽然大多数现有系统通过自然语言, 进行通信，但最近的工作表明，潜在通信, 特别是通过转换器键值(KV) 缓存, 可以提高效率并保留更丰富的任务相关信息。然而，, KV 缓存还对上下文输入, 中间推理状态, 和特定于代理的信息, 进行编码，创建了一个不透明的通道，敏感内容可以通过该通道在代理之间传播，而无需明确的文本披露。为了解决这个,，我们引入了 \textbf{LCGuard} (Latent Communication Guard), 一个在多代理 LLM 系统中基于 KV 的安全潜在通信框架。 LCGuard 将共享 KV 缓存视为潜在工作内存，并在缓存工件跨代理传输之前学习表示级转换。我们通过重建来形式化表示级敏感信息泄漏:，如果对抗性解码器可以从中恢复特定于代理的敏感输入，则共享缓存工件是不安全的。这导致了一种对抗性训练公式，其中对手学习重建敏感输入,，而 LCGuard 学习保留任务相关语义并减少可重建信息的转换。跨多个模型系列和多代理基准的实证评估表明，与标准 KV 共享基线相比，LCGuard 持续降低了基于重建的泄漏和攻击成功率，同时保持了有竞争力的任务性能。
Large language model (LLM)-based multi-agent systems increasingly rely on intermediate communication to coordinate complex tasks. While most existing systems communicate through natural language, recent work shows that latent communication, particularly through transformer key-value (KV) caches, can improve efficiency and preserve richer task-relevant information. However, KV caches also encode contextual inputs, intermediate reasoning states, and agent-specific information, creating an opaque channel through which sensitive content may propagate across agents without explicit textual disclosure. To address this, we introduce \textbf{LCGuard} (Latent Communication Guard), a framework for safe KV-based latent communication in multi-agent LLM systems. LCGuard treats shared KV caches as latent working memory and learns representation-level transformations before cache artifacts are transmitted across agents. We formalize representation-level sensitive information leakage operationally through reconstruction: a shared cache artifact is unsafe if an adversarial decoder can recover agent-specific sensitive inputs from it. This leads to an adversarial training formulation in which the adversary learns to reconstruct sensitive inputs, while LCGuard learns transformations that preserve task-relevant semantics and reduce reconstructable information. Empirical evaluations across multiple model families and multi-agent benchmarks show that LCGuard consistently reduces reconstruction-based leakage and attack success rates while maintaining competitive task performance compared to standard KV-sharing baselines.
科目: 人工智能 (cs.AI); 新兴技术 (cs.ET); 机器学习 (cs.LG); 多代理系统 (cs.MA)
Subjects: Artificial Intelligence (cs.AI); Emerging Technologies (cs.ET); Machine Learning (cs.LG); Multiagent Systems (cs.MA)