S-Agents: self-organizing agents in open-ended environment

Agent organization is a group of agents with a certain structure cooperating for shared goals. During their collaborative process, they autonomously orchestrated workflows without fixed steps by humans. Leveraging large language models (LLMs), autonomous agents have significantly improved, gaining the ability to handle a variety of tasks. In open-ended settings, optimizing collaboration for efficiency and effectiveness demands flexible adjustments. Despite this, current research mainly emphasizes fixed, task-oriented workflows and overlooks agent-centric organizational structures. Drawing inspiration from human organizational behavior, we introduce a self-organizing agent system (S-Agents) with a "tree of agents" structure for dynamic workflow, an "hourglass agent architecture" for balancing information priorities, and a "nonobstructive collaboration" method to allow asynchronous task execution among agents. This structure can autonomously coordinate a group of agents, efficiently addressing the challenges of an open and dynamic environment without human intervention. Our experiments demonstrate that S-Agents proficiently execute collaborative building tasks and resource collection in the Minecraft environment, validating their effectiveness. These authors contributed equally to this work. Li Zhang (lizhangfd@fudan.edu.cn) is the corresponding author with School of Data Science, Fudan University. The fundamental objective of artificial intelligence has long been the development of intelligent autonomous agents with the capacity to operate proficiently in open-ended environments (Weinbaum & Veitas, 2017; Fujita, 2009).