Large Language Models (LLMs) as Agents for Augmented Democracy

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gudiño-Rosero, Jairo, Grandi, Umberto, Hidalgo, César A.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911974096371712
author Gudiño-Rosero, Jairo
Grandi, Umberto
Hidalgo, César A.
author_facet Gudiño-Rosero, Jairo
Grandi, Umberto
Hidalgo, César A.
contents We explore an augmented democracy system built on off-the-shelf LLMs fine-tuned to augment data on citizen's preferences elicited over policies extracted from the government programs of the two main candidates of Brazil's 2022 presidential election. We use a train-test cross-validation setup to estimate the accuracy with which the LLMs predict both: a subject's individual political choices and the aggregate preferences of the full sample of participants. At the individual level, we find that LLMs predict out of sample preferences more accurately than a "bundle rule", which would assume that citizens always vote for the proposals of the candidate aligned with their self-reported political orientation. At the population level, we show that a probabilistic sample augmented by an LLM provides a more accurate estimate of the aggregate preferences of a population than the non-augmented probabilistic sample alone. Together, these results indicates that policy preference data augmented using LLMs can capture nuances that transcend party lines and represents a promising avenue of research for data augmentation.
format Preprint
id arxiv_https___arxiv_org_abs_2405_03452
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Large Language Models (LLMs) as Agents for Augmented Democracy
Gudiño-Rosero, Jairo
Grandi, Umberto
Hidalgo, César A.
Computers and Society
Artificial Intelligence
Computation and Language
We explore an augmented democracy system built on off-the-shelf LLMs fine-tuned to augment data on citizen's preferences elicited over policies extracted from the government programs of the two main candidates of Brazil's 2022 presidential election. We use a train-test cross-validation setup to estimate the accuracy with which the LLMs predict both: a subject's individual political choices and the aggregate preferences of the full sample of participants. At the individual level, we find that LLMs predict out of sample preferences more accurately than a "bundle rule", which would assume that citizens always vote for the proposals of the candidate aligned with their self-reported political orientation. At the population level, we show that a probabilistic sample augmented by an LLM provides a more accurate estimate of the aggregate preferences of a population than the non-augmented probabilistic sample alone. Together, these results indicates that policy preference data augmented using LLMs can capture nuances that transcend party lines and represents a promising avenue of research for data augmentation.
title Large Language Models (LLMs) as Agents for Augmented Democracy
topic Computers and Society
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2405.03452