Can GPT-4 Models Detect Misleading Visualizations?

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Alexander, Jason, Nanda, Priyal, Yang, Kai-Cheng, Sarvghad, Ali
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915296835534848
author Alexander, Jason
Nanda, Priyal
Yang, Kai-Cheng
Sarvghad, Ali
author_facet Alexander, Jason
Nanda, Priyal
Yang, Kai-Cheng
Sarvghad, Ali
contents The proliferation of misleading visualizations online, particularly during critical events like public health crises and elections, poses a significant risk. This study investigates the capability of GPT-4 models (4V, 4o, and 4o mini) to detect misleading visualizations. Utilizing a dataset of tweet-visualization pairs containing various visual misleaders, we test these models under four experimental conditions with different levels of guidance. We show that GPT-4 models can detect misleading visualizations with moderate accuracy without prior training (naive zero-shot) and that performance notably improves when provided with definitions of misleaders (guided zero-shot). However, a single prompt engineering technique does not yield the best results for all misleader types. Specifically, providing the models with misleader definitions and examples (guided few-shot) proves more effective for reasoning misleaders, while guided zero-shot performs better for design misleaders. This study underscores the feasibility of using large vision-language models to detect visual misinformation and the importance of prompt engineering for optimized detection accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2408_12617
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Can GPT-4 Models Detect Misleading Visualizations?
Alexander, Jason
Nanda, Priyal
Yang, Kai-Cheng
Sarvghad, Ali
Computer Vision and Pattern Recognition
Computers and Society
Social and Information Networks
The proliferation of misleading visualizations online, particularly during critical events like public health crises and elections, poses a significant risk. This study investigates the capability of GPT-4 models (4V, 4o, and 4o mini) to detect misleading visualizations. Utilizing a dataset of tweet-visualization pairs containing various visual misleaders, we test these models under four experimental conditions with different levels of guidance. We show that GPT-4 models can detect misleading visualizations with moderate accuracy without prior training (naive zero-shot) and that performance notably improves when provided with definitions of misleaders (guided zero-shot). However, a single prompt engineering technique does not yield the best results for all misleader types. Specifically, providing the models with misleader definitions and examples (guided few-shot) proves more effective for reasoning misleaders, while guided zero-shot performs better for design misleaders. This study underscores the feasibility of using large vision-language models to detect visual misinformation and the importance of prompt engineering for optimized detection accuracy.
title Can GPT-4 Models Detect Misleading Visualizations?
topic Computer Vision and Pattern Recognition
Computers and Society
Social and Information Networks
url https://arxiv.org/abs/2408.12617