FlockVote: LLM-Empowered Agent-Based Modeling for Simulating U.S. Presidential Elections

2025-11-27

AI Summary

Traditional agent-based models (ABMs) suffer from rigid rule specification, while statistical models lack interpretability—limiting their utility for understanding complex sociopolitical decision-making. Method: We propose an LLM-driven, high-fidelity, interpretable agent modeling framework for simulating voter behavior in U.S. presidential elections. Each voter agent is endowed with real-world demographic attributes and dynamically updated policy contexts; individual-level voting decisions are generated via multi-agent generative reasoning, enabling traceable reasoning chains and system-level sensitivity analysis. Contribution/Results: We introduce the novel paradigm of an “LLM-augmented interpretable computational laboratory,” synergizing ABM’s structural rigor with LLMs’ semantic reasoning capabilities. Applied to simulated elections across seven pivotal U.S. states in 2024, our framework reproduces actual vote-share distributions with a mean absolute error <1.2%, while delivering auditable micro-level decision pathways and macro-level stability assessments—establishing a new paradigm for modeling complex social systems.

Technology Category

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

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Modeling complex human behavior, such as voter decisions in national elections, is a long-standing challenge for computational social science. Traditional agent-based models (ABMs) are limited by oversimplified rules, while large-scale statistical models often lack interpretability. We introduce FlockVote, a novel framework that uses Large Language Models (LLMs) to build a"computational laboratory"of LLM agents for political simulation. Each agent is instantiated with a high-fidelity demographic profile and dynamic contextual information (e.g. candidate policies), enabling it to perform nuanced, generative reasoning to simulate a voting decision. We deploy this framework as a testbed on the 2024 U.S. Presidential Election, focusing on seven key swing states. Our simulation's macro-level results successfully replicate the real-world outcome, demonstrating the high fidelity of our"virtual society". The primary contribution is not only the prediction, but also the framework's utility as an interpretable research tool. FlockVote moves beyond black-box outputs, allowing researchers to probe agent-level rationale and analyze the stability and sensitivity of LLM-driven social simulations.

Problem

• Simulate voter decisions using LLM agents
• Improve interpretability of political election models
• Analyze agent-level reasoning in social simulations

Innovation

• LLM agents simulate voting with demographic profiles
• Generative reasoning enables nuanced decision-making in elections
• Framework provides interpretable analysis of simulation stability