Parallelized Planning-Acting for Efficient LLM-based Multi-Agent Systems in Minecraft

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Yaoru, Liu, Shunyu, Zheng, Tongya, Sun, Li, Song, Mingli
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910044337995776
author Li, Yaoru
Liu, Shunyu
Zheng, Tongya
Sun, Li
Song, Mingli
author_facet Li, Yaoru
Liu, Shunyu
Zheng, Tongya
Sun, Li
Song, Mingli
contents Recent advancements in Large Language Model~(LLM)-based Multi-Agent Systems (MAS) have demonstrated remarkable potential for tackling complex decision-making tasks. However, existing frameworks inevitably rely on serialized execution paradigms, where agents must complete sequential LLM planning before taking action. This fundamental constraint severely limits real-time responsiveness and adaptation, which is crucial in dynamic environments with ever-changing scenarios like Minecraft. In this paper, we propose a novel parallelized planning-acting framework for LLM-based MAS, featuring a dual-thread architecture with interruptible execution to enable concurrent planning and acting. Specifically, our framework comprises two core threads: (1) a planning thread driven by a centralized memory system, maintaining synchronization of environmental states and agent communication to support dynamic decision-making; and (2) an acting thread equipped with a comprehensive skill library, enabling automated task execution through recursive decomposition. Extensive experiments on Minecraft demonstrate the effectiveness of the proposed framework.
format Preprint
id arxiv_https___arxiv_org_abs_2503_03505
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Parallelized Planning-Acting for Efficient LLM-based Multi-Agent Systems in Minecraft
Li, Yaoru
Liu, Shunyu
Zheng, Tongya
Sun, Li
Song, Mingli
Artificial Intelligence
Recent advancements in Large Language Model~(LLM)-based Multi-Agent Systems (MAS) have demonstrated remarkable potential for tackling complex decision-making tasks. However, existing frameworks inevitably rely on serialized execution paradigms, where agents must complete sequential LLM planning before taking action. This fundamental constraint severely limits real-time responsiveness and adaptation, which is crucial in dynamic environments with ever-changing scenarios like Minecraft. In this paper, we propose a novel parallelized planning-acting framework for LLM-based MAS, featuring a dual-thread architecture with interruptible execution to enable concurrent planning and acting. Specifically, our framework comprises two core threads: (1) a planning thread driven by a centralized memory system, maintaining synchronization of environmental states and agent communication to support dynamic decision-making; and (2) an acting thread equipped with a comprehensive skill library, enabling automated task execution through recursive decomposition. Extensive experiments on Minecraft demonstrate the effectiveness of the proposed framework.
title Parallelized Planning-Acting for Efficient LLM-based Multi-Agent Systems in Minecraft
topic Artificial Intelligence
url https://arxiv.org/abs/2503.03505