Saved in:
Bibliographic Details
Main Authors: Zhou, Qiangong, Wang, Zhiting, Yao, Mingyou, Liu, Zongyang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.11294
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911106139684864
author Zhou, Qiangong
Wang, Zhiting
Yao, Mingyou
Liu, Zongyang
author_facet Zhou, Qiangong
Wang, Zhiting
Yao, Mingyou
Liu, Zongyang
contents We introduce a new Multi-Agent System (MAS) - Allen, designed to address two core challenges in current MAS design: (1) improve system's policy autonomy, empowering agents to dynamically adapt their behavioral strategies, and (2) achieving the trade-off between collaborative efficiency, task supervision, and human oversight in complex network topologies. Our core insight is to redefine the basic execution unit in the MAS, allowing agents to autonomously form different patterns by combining these units. We have constructed a four-tier state architecture (Task, Stage, Agent, Step) to constrain system behavior from both task-oriented and execution-oriented perspectives. This achieves a unification of topological optimization and controllable progress. Allen grants unprecedented Policy Autonomy, while making a trade-off for the controllability of the collaborative structure. The project code has been open source at: https://github.com/motern88/Allen
format Preprint
id arxiv_https___arxiv_org_abs_2508_11294
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Allen: Rethinking MAS Design through Step-Level Policy Autonomy
Zhou, Qiangong
Wang, Zhiting
Yao, Mingyou
Liu, Zongyang
Multiagent Systems
Computer Vision and Pattern Recognition
We introduce a new Multi-Agent System (MAS) - Allen, designed to address two core challenges in current MAS design: (1) improve system's policy autonomy, empowering agents to dynamically adapt their behavioral strategies, and (2) achieving the trade-off between collaborative efficiency, task supervision, and human oversight in complex network topologies. Our core insight is to redefine the basic execution unit in the MAS, allowing agents to autonomously form different patterns by combining these units. We have constructed a four-tier state architecture (Task, Stage, Agent, Step) to constrain system behavior from both task-oriented and execution-oriented perspectives. This achieves a unification of topological optimization and controllable progress. Allen grants unprecedented Policy Autonomy, while making a trade-off for the controllability of the collaborative structure. The project code has been open source at: https://github.com/motern88/Allen
title Allen: Rethinking MAS Design through Step-Level Policy Autonomy
topic Multiagent Systems
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.11294