Beyond Static Snapshots: Dynamic Modeling and Forecasting of Group-Level Value Evolution with Large Language Models

2026-02-15

AI Summary

This study addresses a critical limitation in current large language model (LLM)-driven social simulations, which typically treat collective values as static and overlook their dynamic evolution under societal change. To bridge this gap, the authors propose the first event-based framework for dynamically predicting group-level value shifts, integrating longitudinal data from multiple waves of the World Values Survey, LLM-generated responses, event embeddings, and time-series modeling to construct a dynamic simulation system for multidimensional populations in the United States and China. The work also introduces the first longitudinal dataset of group-level values. Evaluated across five LLM families, the approach significantly outperforms baselines, achieving performance gains of 30.88% and 33.97% on seen and unseen questions, respectively. Analyses further reveal that U.S. populations exhibit greater value volatility than their Chinese counterparts, with younger cohorts showing heightened sensitivity to external events.

Technology Category

• Humans and AI: Learning Human Values and Preferences
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Social simulation is critical for mining complex social dynamics and supporting data-driven decision making. LLM-based methods have emerged as powerful tools for this task by leveraging human-like social questionnaire responses to model group behaviors. Existing LLM-based approaches predominantly focus on group-level values at discrete time points, treating them as static snapshots rather than dynamic processes. However, group-level values are not fixed but shaped by long-term social changes. Modeling their dynamics is thus crucial for accurate social evolution prediction--a key challenge in both data mining and social science. This problem remains underexplored due to limited longitudinal data, group heterogeneity, and intricate historical event impacts. To bridge this gap, we propose a novel framework for group-level dynamic social simulation by integrating historical value trajectories into LLM-based human response modeling. We select China and the U.S. as representative contexts, conducting stratified simulations across four core sociodemographic dimensions (gender, age, education, income). Using the World Values Survey, we construct a multi-wave, group-level longitudinal dataset to capture historical value evolution, and then propose the first event-based prediction method for this task, unifying social events, current value states, and group attributes into a single framework. Evaluations across five LLM families show substantial gains: a maximum 30.88\% improvement on seen questions and 33.97\% on unseen questions over the Vanilla baseline. We further find notable cross-group heterogeneity: U.S. groups are more volatile than Chinese groups, and younger groups in both countries are more sensitive to external changes. These findings advance LLM-based social simulation and provide new insights for social scientists to understand and predict social value changes.

Problem

• dynamic social simulation
• group-level value evolution
• longitudinal modeling
• LLM-based social modeling
• value forecasting

Innovation

• dynamic social simulation
• large language models
• value evolution
• longitudinal modeling
• event-based prediction