Intelligent Computing Social Modeling and Methodological Innovations in Political Science in the Era of Large Language Models

Abstract

The recent wave of artificial intelligence, epitomized by large language models (LLMs),has presented opportunities and challenges for methodological innovation in political science,sparking discussions on a potential paradigm shift in the social sciences. However, how can weunderstand the impact of LLMs on knowledge production and paradigm transformation in thesocial sciences from a comprehensive perspective that integrates technology and methodology? What are LLMs' specific applications and representative innovative methods in political scienceresearch? These questions, particularly from a practical methodological standpoint, remainunderexplored. This paper proposes the "Intelligent Computing Social Modeling" (ICSM) methodto address these issues by clarifying the critical mechanisms of LLMs. ICSM leverages thestrengths of LLMs in idea synthesis and action simulation, advancing intellectual exploration inpolitical science through "simulated social construction" and "simulation validation." Bysimulating the U.S. presidential election, this study empirically demonstrates the operationalpathways and methodological advantages of ICSM. By integrating traditional social scienceparadigms, ICSM not only enhances the quantitative paradigm's capability to apply big data toassess the impact of factors but also provides qualitative paradigms with evidence for socialmechanism discovery at the individual level, offering a powerful tool that balances interpretabilityand predictability in social science research. The findings suggest that LLMs will drivemethodological innovation in political science through integration and improvement rather thandirect substitution.

37 pages, 11 figures, 3 tables. J OF CHIN POLIT SCI (2025)

References in corpus (15)

• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• Scaling Laws for Neural Language Models
• Emergent Abilities of Large Language Models
• Self-Consistency Improves Chain of Thought Reasoning in Language Models
• Out of One, Many: Using Language Models to Simulate Human Samples
• The Rise and Potential of Large Language Model Based Agents: A Survey
• Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models
• Large Language Model as Attributed Training Data Generator: A Tale of Diversity and Bias
• ART: Automatic multi-step reasoning and tool-use for large language models
• Exploring Large Language Models for Communication Games: An Empirical Study on Werewolf
• War and Peace (WarAgent): Large Language Model-based Multi-Agent Simulation of World Wars
• Reasoning with Language Model is Planning with World Model
• How Far Are We on the Decision-Making of LLMs? Evaluating LLMs' Gaming Ability in Multi-Agent Environments
• CompeteAI: Understanding the Competition Dynamics in Large Language Model-based Agents
• Random Silicon Sampling: Simulating Human Sub-Population Opinion Using a Large Language Model Based on Group-Level Demographic Information