AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors

TL;DR

AgentVerse introduces a multi-agent framework that enables collaborative, dynamic group formation with emergent social behaviors, outperforming individual agents in complex tasks, inspired by human group dynamics.

Contribution

This work presents a novel multi-agent framework that dynamically adjusts group composition and explores emergent social behaviors to enhance collaborative task performance.

Findings

Multi-agent groups outperform single agents in tasks.

Emergent social behaviors influence group collaboration.

Strategies can leverage positive behaviors and mitigate negatives.

Abstract

Autonomous agents empowered by Large Language Models (LLMs) have undergone significant improvements, enabling them to generalize across a broad spectrum of tasks. However, in real-world scenarios, cooperation among individuals is often required to enhance the efficiency and effectiveness of task accomplishment. Hence, inspired by human group dynamics, we propose a multi-agent framework \framework that can collaboratively and dynamically adjust its composition as a greater-than-the-sum-of-its-parts system. Our experiments demonstrate that \framework framework can effectively deploy multi-agent groups that outperform a single agent. Furthermore, we delve into the emergence of social behaviors among individual agents within a group during collaborative task accomplishment. In view of these behaviors, we discuss some possible strategies to leverage positive ones and mitigate negative ones for improving the collaborative potential of multi-agent groups. Our codes for \framework will soon be released at \url{https://github.com/OpenBMB/AgentVerse}.