Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Ruizhe, Jiang, Xinke, Yang, Zhibang, Zhang, Zhixin, Gao, Jiaran, Xiao, Yuzhen, Lai, Hongbin, Chu, Xu, Zhao, Junfeng, Wang, Yasha
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2601.05890
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917192456470528
author Zhang, Ruizhe
Jiang, Xinke
Yang, Zhibang
Zhang, Zhixin
Gao, Jiaran
Xiao, Yuzhen
Lai, Hongbin
Chu, Xu
Zhao, Junfeng
Wang, Yasha
author_facet Zhang, Ruizhe
Jiang, Xinke
Yang, Zhibang
Zhang, Zhixin
Gao, Jiaran
Xiao, Yuzhen
Lai, Hongbin
Chu, Xu
Zhao, Junfeng
Wang, Yasha
contents Multi-agent systems based on large language models, particularly centralized architectures, have recently shown strong potential for complex and knowledge-intensive tasks. However, central agents often suffer from unstable long-horizon collaboration due to the lack of memory management, leading to context bloat, error accumulation, and poor cross-task generalization. To address both task-level memory inefficiency and the inability to reuse coordination experience, we propose StackPlanner, a hierarchical multi-agent framework with explicit memory control. StackPlanner addresses these challenges by decoupling high-level coordination from subtask execution with active task-level memory control, and by learning to retrieve and exploit reusable coordination experience via structured experience memory and reinforcement learning. Experiments on multiple deep-search and agent system benchmarks demonstrate the effectiveness of our approach in enabling reliable long-horizon multi-agent collaboration.
format Preprint
id arxiv_https___arxiv_org_abs_2601_05890
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle StackPlanner: A Centralized Hierarchical Multi-Agent System with Task-Experience Memory Management
Zhang, Ruizhe
Jiang, Xinke
Yang, Zhibang
Zhang, Zhixin
Gao, Jiaran
Xiao, Yuzhen
Lai, Hongbin
Chu, Xu
Zhao, Junfeng
Wang, Yasha
Artificial Intelligence
Multi-agent systems based on large language models, particularly centralized architectures, have recently shown strong potential for complex and knowledge-intensive tasks. However, central agents often suffer from unstable long-horizon collaboration due to the lack of memory management, leading to context bloat, error accumulation, and poor cross-task generalization. To address both task-level memory inefficiency and the inability to reuse coordination experience, we propose StackPlanner, a hierarchical multi-agent framework with explicit memory control. StackPlanner addresses these challenges by decoupling high-level coordination from subtask execution with active task-level memory control, and by learning to retrieve and exploit reusable coordination experience via structured experience memory and reinforcement learning. Experiments on multiple deep-search and agent system benchmarks demonstrate the effectiveness of our approach in enabling reliable long-horizon multi-agent collaboration.
title StackPlanner: A Centralized Hierarchical Multi-Agent System with Task-Experience Memory Management
topic Artificial Intelligence
url https://arxiv.org/abs/2601.05890