Learning dynamics from online-offline systems of LLM agents

2026-02-26

AI Summary

This study investigates how real-world events and personality traits of large language model (LLM) agents jointly shape online information diffusion, revealing their connections to system dynamics and real-world instability. By constructing an agent-based model in which LLM agents endowed with 32 distinct personality traits disseminate news about armed conflicts across a k-regular random network, and by deriving a system of ordinary differential equations via mean-field approximation, this work pioneers the integration of LLM personality and event type into diffusion modeling. The findings demonstrate that, despite the high complexity of individual behaviors, the aggregate propagation dynamics can be effectively captured by a simplified SI model with only two transmission rates, highlighting an underlying simplicity in complex information diffusion processes.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Modeling other Agents
• Natural Language Processing: (Large) Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Online information is increasingly linked to real-world instability, especially as automated accounts and LLM-based agents help spread and amplify news. In this work, we study how information spreads on networks of Large Language Models (LLMs) using mathematical models. We investigate how different types of offline events, along with the"personalities"assigned to the LLMs, affect the network dynamics of online information spread of the events among the LLMs. We introduce two models: 1) a stochastic agent-based network model and 2) a system of differential equations arising from a mean-field approximation to the agent-based model. We fit these models to simulations of the spread of armed-conflict news on social media, using LLM agents each with one of 32 personality trait profiles on k-regular random networks. Our results indicate that, despite the complexity of the news events, personalities, and LLM behaviors, the overall dynamics of the system are well described by a Susceptible-Infected (SI) type model with two transmission rates.

Problem

• LLM agents
• information spread
• network dynamics
• personality traits
• online-offline systems

Innovation

• LLM agents
• personality traits
• information diffusion
• agent-based modeling
• mean-field approximation