Can Agents Spontaneously Form a Society? Introducing a Novel Architecture for Generative Multi-Agents to Elicit Social Emergence

TL;DR

This paper presents ITCMA-S, a novel multi-agent architecture that fosters social interactions and emergent social structures like hierarchies and cliques in a simulated environment.

Contribution

The introduction of ITCMA-S and LTRHA frameworks enables generative agents to engage in social behaviors and form complex social structures.

Findings

Agents can recognize and adapt to social cues.

Formation of hierarchies and cliques among agents.

Effective exploration and information sharing in social settings.

Abstract

Generative agents have demonstrated impressive capabilities in specific tasks, but most of these frameworks focus on independent tasks and lack attention to social interactions. We introduce a generative agent architecture called ITCMA-S, which includes a basic framework for individual agents and a framework called LTRHA that supports social interactions among multi-agents. This architecture enables agents to identify and filter out behaviors that are detrimental to social interactions, guiding them to choose more favorable actions. We designed a sandbox environment to simulate the natural evolution of social relationships among multiple identity-less agents for experimental evaluation. The results showed that ITCMA-S performed well on multiple evaluation indicators, demonstrating its ability to actively explore the environment, recognize new agents, and acquire new information through continuous actions and dialogue. Observations show that as agents establish connections with each other, they spontaneously form cliques with internal hierarchies around a selected leader and organize collective activities.