Σ-Mem：面向基于大语言模型的多智能体系统的在线可靠性记忆

摘要

记忆对于长时程大语言模型（LLM）智能体至关重要，然而现有记忆系统主要保存交互内容，而非建模哪些智能体值得信任及其适用条件。这一局限在多智能体系统中尤为突出——中心模型可能无法直接验证看似合理或相互关联的对等体响应。我们提出Σ-Mem，一种在线可靠性记忆，它记录个体对等体的历史能力证据以及整个对等体集合中的对等体间关系证据。两类证据均以实对称状态维护，并根据决策后的正确性反馈进行更新。根据Weyl不等式，每次事件级更新引起的谱变化是有界的，从而无需重新训练底层模型即可实现稳定的在线自适应。Σ-Mem提供通用的写入—读取接口：同一记忆可用于中心模型的残差引导、无响应式对等体路由或可靠性加权投票。在五个Qwen系列模型上，Σ-Mem能够适应反事实的可靠性变化，并泛化至未见过的对等体和任务领域。在完整分布外（OOD）评估集上，直接记忆读取在性能上也优于多数投票和最佳固定对等体。此外，随着可获得更多正确性反馈，性能持续提升，表明Σ-Mem逐步积累了可操作的可靠性信息。这些结果确立了可靠性记忆作为基于LLM的多智能体系统中自适应协调的可复用基础。

English

Memory is central to long-horizon LLM agents, yet existing memory systems primarily preserve interaction content rather than modeling which agents can be trusted and under what conditions. This limitation is particularly important in multi-agent systems, where a central model may be unable to directly verify plausible or correlated peer responses. We introduce Σ-Mem, an online reliability memory that records historical competence evidence for individual peers and peer relationship evidence across the peer set. Both forms of evidence are maintained as real symmetric states and updated from post-decision correctness feedback. By Weyl's inequality, the spectral change caused by each event-level update is bounded, enabling stable online adaptation without retraining the underlying models. Σ-Mem provides a general write-and-read interface: the same memory can be used for residual steering of a central model, response-free peer routing, or reliability-weighted voting. Across five Qwen-family models, Σ-Mem adapts to counterfactual reliability shifts and generalizes to unseen peers and task domains. Direct memory readouts also outperform majority voting and the best fixed peer over the full OOD evaluation set. Moreover, performance improves consistently as more correctness feedback becomes available, indicating that Σ-Mem progressively accumulates actionable reliability information. These results establish reliability memory as a reusable foundation for adaptive coordination in LLM-based multi-agent systems.