A Large-Scale Simulation on Large Language Models for Decision-Making in Political Science

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yu, Chenxiao, Ye, Jinyi, Li, Yuangang, Li, Zheng, Ferrara, Emilio, Hu, Xiyang, Zhao, Yue
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916682253991936
author Yu, Chenxiao
Ye, Jinyi
Li, Yuangang
Li, Zheng
Ferrara, Emilio
Hu, Xiyang
Zhao, Yue
author_facet Yu, Chenxiao
Ye, Jinyi
Li, Yuangang
Li, Zheng
Ferrara, Emilio
Hu, Xiyang
Zhao, Yue
contents While LLMs have demonstrated remarkable capabilities in text generation and reasoning, their ability to simulate human decision-making -- particularly in political contexts -- remains an open question. However, modeling voter behavior presents unique challenges due to limited voter-level data, evolving political landscapes, and the complexity of human reasoning. In this study, we develop a theory-driven, multi-step reasoning framework that integrates demographic, temporal and ideological factors to simulate voter decision-making at scale. Using synthetic personas calibrated to real-world voter data, we conduct large-scale simulations of recent U.S. presidential elections. Our method significantly improves simulation accuracy while mitigating model biases. We examine its robustness by comparing performance across different LLMs. We further investigate the challenges and constraints that arise from LLM-based political simulations. Our work provides both a scalable framework for modeling political decision-making behavior and insights into the promise and limitations of using LLMs in political science research.
format Preprint
id arxiv_https___arxiv_org_abs_2412_15291
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Large-Scale Simulation on Large Language Models for Decision-Making in Political Science
Yu, Chenxiao
Ye, Jinyi
Li, Yuangang
Li, Zheng
Ferrara, Emilio
Hu, Xiyang
Zhao, Yue
Computation and Language
Social and Information Networks
While LLMs have demonstrated remarkable capabilities in text generation and reasoning, their ability to simulate human decision-making -- particularly in political contexts -- remains an open question. However, modeling voter behavior presents unique challenges due to limited voter-level data, evolving political landscapes, and the complexity of human reasoning. In this study, we develop a theory-driven, multi-step reasoning framework that integrates demographic, temporal and ideological factors to simulate voter decision-making at scale. Using synthetic personas calibrated to real-world voter data, we conduct large-scale simulations of recent U.S. presidential elections. Our method significantly improves simulation accuracy while mitigating model biases. We examine its robustness by comparing performance across different LLMs. We further investigate the challenges and constraints that arise from LLM-based political simulations. Our work provides both a scalable framework for modeling political decision-making behavior and insights into the promise and limitations of using LLMs in political science research.
title A Large-Scale Simulation on Large Language Models for Decision-Making in Political Science
topic Computation and Language
Social and Information Networks
url https://arxiv.org/abs/2412.15291