LLM-Guided Communication for Cooperative Multi-Agent Reinforcement Learning

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Bae, Sangjun, Park, Yisak, Lee, Sanghyeon, Han, Seungyul
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916071804502016
author Bae, Sangjun
Park, Yisak
Lee, Sanghyeon
Han, Seungyul
author_facet Bae, Sangjun
Park, Yisak
Lee, Sanghyeon
Han, Seungyul
contents Communication is a key component in multi-agent reinforcement learning (MARL) for mitigating partial observability, yet prior approaches often rely on inefficient information exchange or fail to transmit sufficient state information. To address this, we propose LLM-driven Multi-Agent Communication (LMAC), which leverages an LLM's reasoning capability to design a communication protocol that enables all agents to reconstruct the underlying state as accurately and uniformly as possible. LMAC iteratively refines the protocol using an explicit state-awareness criterion, improving state recovery while narrowing differences in agents' knowledge. Experiments on diverse MARL benchmarks show that LMAC improves state reconstruction across agents and yields substantial performance gains over prior communication baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2605_18077
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle LLM-Guided Communication for Cooperative Multi-Agent Reinforcement Learning
Bae, Sangjun
Park, Yisak
Lee, Sanghyeon
Han, Seungyul
Artificial Intelligence
Machine Learning
Multiagent Systems
Communication is a key component in multi-agent reinforcement learning (MARL) for mitigating partial observability, yet prior approaches often rely on inefficient information exchange or fail to transmit sufficient state information. To address this, we propose LLM-driven Multi-Agent Communication (LMAC), which leverages an LLM's reasoning capability to design a communication protocol that enables all agents to reconstruct the underlying state as accurately and uniformly as possible. LMAC iteratively refines the protocol using an explicit state-awareness criterion, improving state recovery while narrowing differences in agents' knowledge. Experiments on diverse MARL benchmarks show that LMAC improves state reconstruction across agents and yields substantial performance gains over prior communication baselines.
title LLM-Guided Communication for Cooperative Multi-Agent Reinforcement Learning
topic Artificial Intelligence
Machine Learning
Multiagent Systems
url https://arxiv.org/abs/2605.18077