MASim: Multilingual Agent-Based Simulation for Social Science

TL;DR

MASim is a pioneering multilingual agent-based simulation framework that models cross-lingual social interactions, enabling analysis of public opinion and media influence across diverse cultures.

Contribution

It introduces the first multilingual simulation platform supporting multi-turn interactions among diverse sociolinguistic agents, and provides the MAPS benchmark for sociocultural analysis.

Findings

MASim reproduces sociocultural phenomena.

Multilingual simulation enhances understanding of cross-cultural dynamics.

The framework supports scalable computational social science.

Abstract

Multi-agent role-playing has recently shown promise for studying social behavior with language agents, but existing simulations are mostly monolingual and fail to model cross-lingual interaction, an essential property of real societies. We introduce MASim, the first multilingual agent-based simulation framework that supports multi-turn interaction among generative agents with diverse sociolinguistic profiles. MASim offers two key analyses: (i) global public opinion modeling, by simulating how attitudes toward open-domain hypotheses evolve across languages and cultures, and (ii) media influence and information diffusion, via autonomous news agents that dynamically generate content and shape user behavior. To instantiate simulations, we construct the MAPS benchmark, which combines survey questions and demographic personas drawn from global population distributions. Experiments on calibration, sensitivity, consistency, and cultural case studies show that MASim reproduces sociocultural phenomena and highlights the importance of multilingual simulation for scalable, controlled computational social science.