MASim: Multilingual Agent-Based Simulation for Social Science
This work addresses the need for scalable and controlled computational social science by enabling multilingual simulations, though it appears incremental as it extends existing agent-based methods to a multilingual context.
The paper tackles the problem of monolingual limitations in agent-based simulations for social science by introducing MASim, a multilingual framework that models cross-lingual interactions and reproduces sociocultural phenomena, as demonstrated through experiments on calibration, sensitivity, and consistency.
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.