arXiv 2603.060072026-03cs.CLcs.AIcs.MA

MASFactory: 一种基于图的框架，用于通过Vibe图谱编排基于大语言模型的多智能体系统

MASFactory: A Graph-centric Framework for Orchestrating LLM-Based Multi-Agent Systems with Vibe Graphing

Beijing University of Posts and Telecommunications（北京邮电大学）；Shanghai Jiao Tong University（上海交通大学）

Yang Liu, Jinxuan Cai, Yishen Li, Qi Meng, Zedi Liu, Xin Li, Chen Qian, Chuan Shi, Cheng Yang

备注：: Accepted to the ACL 2026 Demo Track. Camera-ready version. 10 pages, 6 figures. Code and documentation are available at: https://github.com/BUPT-GAMMA/MASFactory

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出MASFactory，一种基于图的框架，用于通过Vibe图谱编排基于大语言模型的多智能体系统，解决了现有框架在实现复杂图工作流时需要大量手动工作、重用性差和难以整合异构外部上下文源的问题。

摘要：

基于大语言模型的（LLM-based）多智能体系统（MAS）越来越多地被用于通过角色专业化和协作扩展智能体问题解决。MAS工作流可以自然地建模为有向计算图，其中节点执行智能体或子工作流，边编码依赖性和消息传递。然而，目前框架在实现复杂图工作流时仍然需要大量的手动工作，提供有限的重用性，并使整合异构外部上下文源变得困难。为克服这些限制，我们提出了MASFactory，一种用于编排基于大语言模型的MAS的基于图的框架。它引入了Vibe图谱，一种人机交互的方法，将自然语言意图编译成可编辑的工作流规范，然后编译成可执行的图。此外，该框架提供了可重用的组件、技能支持、多模态消息处理和可插拔的上下文整合，以及用于拓扑预览、运行时跟踪和人机交互的可视化工具。我们在七个公开基准上评估了MASFactory，验证了代表性MAS方法的再生产一致性以及Vibe图谱的有效性。我们的代码（https://github.com/BUPT-GAMMA/MASFactory，Apache-2.0许可）和视频演示（https://youtu.be/ANynzVfY32k）均已公开。

英文摘要：

Large language model-based (LLM-based) multi-agent systems (MAS) are increasingly used to extend agentic problem solving via role specialization and collaboration. MAS workflows can be naturally modeled as directed computation graphs, where nodes execute agents or sub-workflows and edges encode dependencies and message passing. However, implementing complex graph workflows in current frameworks still requires substantial manual effort, offers limited reuse, and makes it difficult to integrate heterogeneous external context sources. To overcome these limitations, we present MASFactory, a graph-centric framework for orchestrating LLM-based MAS. It introduces Vibe Graphing, a human-in-the-loop approach that compiles natural-language intent into an editable workflow specification and then into an executable graph. In addition, the framework provides reusable components, skill support, multimodal message handling, and pluggable context integration, as well as a visualizer for topology preview, runtime tracing, and human-in-the-loop interaction. We evaluate MASFactory on seven public benchmarks, validating both reproduction consistency for representative MAS methods and the effectiveness of Vibe Graphing. Our code (https://github.com/BUPT-GAMMA/MASFactory, licensed under Apache-2.0) and video demonstration (https://youtu.be/ANynzVfY32k) are publicly available.

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。