[提交于 2026 年 7 月 3 日]

标题：面向智能体执行业务流程的组织记忆

查看 PDF HTML（实验性）

摘要：基于 LLM 的智能体为超越规则系统的局限、实现业务流程执行的自动化提供了新的机遇。然而，通用 LLM 缺乏可靠执行所需的组织特定知识，而这些知识通常散落在以人为主体的各类制品中，例如政策、流程模型和标准操作规程。虽然这些知识在技术上可以被编码到单个提示或面向特定智能体的检索设置中，但这种方式在企业中无法规模化，因为它会导致知识孤岛与规则重复，并使得跨智能体的一致更新和学习变得困难。我们认为，这需要为智能体执行业务流程构建组织记忆：即一个关于工作应如何执行的、不断演化的组织特定程序性知识的共享、受治理且可被智能体消费的参考层。我们推导了这种记忆的需求，提出了一套用于其策展与消费的架构，并在基于采购场景的概念验证中展示了其有效性。

提交历史

来自：Lukas Kirchdorfer [

查看邮箱

]

[v1]

2026 年 7 月 3 日，星期五 11:40:10 UTC（1,118 KB）

[Submitted on 3 Jul 2026]

Title:Organizational Memory for Agentic Business Process Execution

View a PDF of the paper titled Organizational Memory for Agentic Business Process Execution, by Lukas Kirchdorfer and 5 other authors

View PDF HTML (experimental)

Abstract:LLM-based agents offer new opportunities for automating business process execution beyond the limits of rule-based systems. However, general-purpose LLMs lack the organization-specific knowledge required for reliable execution, which is typically fragmented across human-oriented artifacts such as policies, process models, and standard operating procedures. While such knowledge can technically be encoded in individual prompts or agent-specific retrieval setups, this approach does not scale in enterprises, as it gives rise to knowledge silos and rule duplicates, and makes consistent updates and learning across agents difficult. We argue that this calls for an organizational memory for agentic business process execution: a shared, governed, and agent-consumable reference layer of evolving organization-specific procedural knowledge about how work should be executed. We derive requirements for such a memory, propose an architecture for its curation and consumption, and demonstrate its effectiveness in a proof-of-concept based on a procurement scenario.

Submission history

From: Lukas Kirchdorfer [

view email

]

[v1]

Fri, 3 Jul 2026 11:40:10 UTC (1,118 KB)