Network Formation and Dynamics Among Multi-LLMs

2024-02-16

arXiv.org

AI Summary

This study investigates whether large language models (LLMs) can emulate the dynamic mechanisms underlying human social network formation. We propose a multi-agent interaction framework integrating state-of-the-art LLMs—including GPT, Claude, and Llama—and conduct controlled experiments on three empirically grounded network topologies: Facebook, telephone communication, and workplace interactions. Using quantitative metrics—preferential attachment, triadic closure, homophily, community structure, and small-worldness—we systematically demonstrate that LLM-based agent populations spontaneously generate key structural properties observed in human networks. Crucially, we identify scenario-adaptive heterogeneity in network formation: for instance, in the workplace setting, LLM agents accurately reproduce higher-order connectivity patterns and structural heterogeneity. These findings establish a novel, interpretable, and generalizable paradigm for synthetic network generation and agent-based modeling (ABM), bridging foundational network science with foundation model capabilities.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Social networks fundamentally shape human opinions, behaviors, and the dissemination of information. As large language models (LLMs) like GPT, Claude, and Llama increasingly integrate into social and professional settings, understanding their behavior in the context of social interactions and network formation becomes essential. This study develops a framework to systematically examine whether the network formation behaviors of multiple LLMs approximate certain aspects of human network dynamics. By simulating interactions among LLM agents across various model families, we observe that these models consistently exhibit key patterns associated with social network principles including preferential attachment, triadic closure, homophily, community structure, and the small-world phenomenon when forming networks. Moreover, LLMs adapt their network formation strategies based on each network's characteristics, reflecting the context-dependent nature of human behavior: in Facebook networks, they prioritize triadic closure and homophily, mirroring close-knit friendships; in phone networks, homophily and preferential attachment dominate, capturing personal and professional connections, while in employment networks, LLMs favor heterophily and high-degree connections, aligning with career advancement dynamics. These results open new avenues for using LLMs in network science research, with potential applications in agent-based modeling and synthetic network generation.

Problem

• Studying LLM agents' network formation behaviors
• Benchmarking LLMs against human network decisions
• Investigating micro and macro-level network dynamics reproduction

Innovation

• Multi-LLM agent network framework
• Benchmarking against human network decisions
• Context-aware social principle adaptation