Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Survey on LLM-based Multi-Agent System: Recent Advances and New Frontiers in Application

Shuaihang Chen, Yuanxing Liu, Wei Han, Weinan Zhang, Ting Liu

TL;DR

This survey addresses the rapid growth of LLM-based Multi-Agent Systems (LLM-MAS) by offering an application-centric taxonomy and a consolidated view of recent advances. It frames LLM-MAS around core components—generative agents and environment—and surveys their use in solving complex tasks, simulating specific scenarios, and evaluating generative agents, supported by open resources and benchmarks. Key contributions include a taxonomy focused on application purpose, a compilation of available frameworks and datasets, and a structured discussion of challenges spanning alignment, efficiency, and evaluation. The work highlights practical implications for building scalable, realistic, and evaluable LLM-MAS, identifying concrete directions for future research and benchmarking efforts.

Abstract

LLM-based Multi-Agent Systems ( LLM-MAS ) have become a research hotspot since the rise of large language models (LLMs). However, with the continuous influx of new related works, the existing reviews struggle to capture them comprehensively. This paper presents a comprehensive survey of these studies. We first discuss the definition of LLM-MAS, a framework encompassing much of previous work. We provide an overview of the various applications of LLM-MAS in (i) solving complex tasks, (ii) simulating specific scenarios, and (iii) evaluating generative agents. Building on previous studies, we also highlight several challenges and propose future directions for research in this field.

A Survey on LLM-based Multi-Agent System: Recent Advances and New Frontiers in Application

TL;DR

This survey addresses the rapid growth of LLM-based Multi-Agent Systems (LLM-MAS) by offering an application-centric taxonomy and a consolidated view of recent advances. It frames LLM-MAS around core components—generative agents and environment—and surveys their use in solving complex tasks, simulating specific scenarios, and evaluating generative agents, supported by open resources and benchmarks. Key contributions include a taxonomy focused on application purpose, a compilation of available frameworks and datasets, and a structured discussion of challenges spanning alignment, efficiency, and evaluation. The work highlights practical implications for building scalable, realistic, and evaluable LLM-MAS, identifying concrete directions for future research and benchmarking efforts.

Abstract

LLM-based Multi-Agent Systems ( LLM-MAS ) have become a research hotspot since the rise of large language models (LLMs). However, with the continuous influx of new related works, the existing reviews struggle to capture them comprehensively. This paper presents a comprehensive survey of these studies. We first discuss the definition of LLM-MAS, a framework encompassing much of previous work. We provide an overview of the various applications of LLM-MAS in (i) solving complex tasks, (ii) simulating specific scenarios, and (iii) evaluating generative agents. Building on previous studies, we also highlight several challenges and propose future directions for research in this field.

Paper Structure

This paper contains 20 sections, 1 figure, 3 tables.

Table of Contents

  1. Introduction
  2. Core Components of LLM-MAS
  3. Generative Agents
  4. Environment
  5. LLM-MAS for Solving Complex Tasks
  6. Representative LLM-MAS for Solving Complex Tasks
  7. Resources of LLM-MAS for Solving Complex Tasks
  8. Evaluation of LLM-MAS for solving complex task
  9. LLM-MAS for Simulating Specific Scenarios
  10. Representative LLM-MAS for Simulating Specific Scenarios
  11. Resources for LLM-MAS simulation
  12. Evaluation of LLM-MAS simulation
  13. LLM-MAS for Evaluating Generative Agents
  14. Representative LLM-MAS for Evaluating Generative Agents
  15. Resources of LLM-MAS for evaluations
  16. ...and 5 more sections

Figures (1)

  • Figure 1: Overview of the application framework and relationship of LLM-MAS, generative agent, and LLM. Dashed-bordered right-angled rectangles represent content aligned with previous surveys, while rounded rectangles indicate original contributions introduced in this study.