arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.17787cs.AIcs.CLcs.MA
收藏

受控内存:多智能体工作流的生产架构

Governed Memory: A Production Architecture for Multi-Agent Workflows

Hamed Taheri

AI总结:

本文提出受控内存架构,通过共享内存和治理层解决多智能体工作流中的内存治理问题,提升系统精度与可靠性。

摘要:

企业AI在工作流中部署多个自主智能体节点,每个节点处理相同实体但无共享内存和共同治理。本文识别出五个结构性挑战:跨智能体工作流的内存孤岛;跨团队和工具的治理碎片化;无法被下游系统使用的无结构内存;自主多步骤执行中的冗余上下文交付;以及无反馈循环导致的静默质量退化。我们提出受控内存,通过四种机制解决这一缺口:双内存模型结合开放式原子事实与模式强制类型属性;分层治理路由与渐进上下文交付;反射受限检索与实体范围隔离;以及闭环模式生命周期与AI辅助编写和自动化属性细化。我们通过受控实验(N=250,五种内容类型)验证每个机制:99.6%的事实回忆率配合互补双模覆盖;92%的治理路由精度;50%的token减少率;500次对抗查询中零跨实体泄漏;100%的对抗治理合规性;以及每实体约七次受控内存的输出质量饱和。在LoCoMo基准测试中,架构达到74.8%的整体准确率,证实治理和模式强制不降低检索质量。该系统已在Personize.ai生产环境中部署。

英文摘要:

Enterprise AI deploys dozens of autonomous agent nodes across workflows, each acting on the same entities with no shared memory and no common governance. We identify five structural challenges arising from this memory governance gap: memory silos across agent workflows; governance fragmentation across teams and tools; unstructured memories unusable by downstream systems; redundant context delivery in autonomous multi-step executions; and silent quality degradation without feedback loops. We present Governed Memory, a shared memory and governance layer addressing this gap through four mechanisms: a dual memory model combining open-set atomic facts with schema-enforced typed properties; tiered governance routing with progressive context delivery; reflection-bounded retrieval with entity-scoped isolation; and a closed-loop schema lifecycle with AI-assisted authoring and automated per-property refinement. We validate each mechanism through controlled experiments (N=250, five content types): 99.6% fact recall with complementary dual-modality coverage; 92% governance routing precision; 50% token reduction from progressive delivery; zero cross-entity leakage across 500 adversarial queries; 100% adversarial governance compliance; and output quality saturation at approximately seven governed memories per entity. On the LoCoMo benchmark, the architecture achieves 74.8% overall accuracy, confirming that governance and schema enforcement impose no retrieval quality penalty. The system is in production at Personize.ai.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑