Summary
Social networks are complex systems where people interact and influence each other over time. This paper reviews how large language models (LLMs), which understand and generate human language, are used to better mimic how these social networks form and change. The authors organize existing work into models that create network connections and those that simulate how opinions and information spread. While LLMs allow for more realistic and detailed social interactions by using natural language, they also face problems like biases and sensitivity to how questions are phrased. The paper highlights challenges and suggests directions for improving LLMs in social network research.
Large language modelsSocial networksNetwork generationDynamic social processesOpinion dynamicsInformation diffusionRumor propagationSocial biasPrompt sensitivity
Authors
Shikha Mallick, Alex Thomo, Akrati Saxena
Abstract
Large language models (LLMs) are rapidly emerging as a new paradigm for modeling social networks by representing users and their relationships and interactions through natural language. Unlike classical network models or deep learning approaches, LLMs can simulate context-aware social behavior and language-driven interactions, enabling more realistic modeling of network formation and dynamic social processes. However, existing studies are scattered across different research communities and lack a unified perspective. This survey presents the first comprehensive review of LLMs for social network modeling by organizing the literature into two broad categories: network generative models and dynamic process models. Network generative models are further classified into selection-based and interaction-based approaches, while dynamic process models are categorized into opinion dynamics, information diffusion, and rumor propagation, each with their underlying modeling mechanisms. LLMs enable rich textual social interactions and decision-making, but they also exhibit many limitations, including inherent social biases and prompt sensitivity. We outline these open research challenges and discuss future directions in LLM-based social network modeling.