Automated Bias Assessment in AI-Generated Educational Content Using CEAT Framework

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Peng, Jingyang, Shen, Wenyuan, Rao, Jiarui, Lin, Jionghao
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916742988562432
author Peng, Jingyang
Shen, Wenyuan
Rao, Jiarui
Lin, Jionghao
author_facet Peng, Jingyang
Shen, Wenyuan
Rao, Jiarui
Lin, Jionghao
contents Recent advances in Generative Artificial Intelligence (GenAI) have transformed educational content creation, particularly in developing tutor training materials. However, biases embedded in AI-generated content--such as gender, racial, or national stereotypes--raise significant ethical and educational concerns. Despite the growing use of GenAI, systematic methods for detecting and evaluating such biases in educational materials remain limited. This study proposes an automated bias assessment approach that integrates the Contextualized Embedding Association Test with a prompt-engineered word extraction method within a Retrieval-Augmented Generation framework. We applied this method to AI-generated texts used in tutor training lessons. Results show a high alignment between the automated and manually curated word sets, with a Pearson correlation coefficient of r = 0.993, indicating reliable and consistent bias assessment. Our method reduces human subjectivity and enhances fairness, scalability, and reproducibility in auditing GenAI-produced educational content.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12718
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Automated Bias Assessment in AI-Generated Educational Content Using CEAT Framework
Peng, Jingyang
Shen, Wenyuan
Rao, Jiarui
Lin, Jionghao
Computation and Language
Human-Computer Interaction
Recent advances in Generative Artificial Intelligence (GenAI) have transformed educational content creation, particularly in developing tutor training materials. However, biases embedded in AI-generated content--such as gender, racial, or national stereotypes--raise significant ethical and educational concerns. Despite the growing use of GenAI, systematic methods for detecting and evaluating such biases in educational materials remain limited. This study proposes an automated bias assessment approach that integrates the Contextualized Embedding Association Test with a prompt-engineered word extraction method within a Retrieval-Augmented Generation framework. We applied this method to AI-generated texts used in tutor training lessons. Results show a high alignment between the automated and manually curated word sets, with a Pearson correlation coefficient of r = 0.993, indicating reliable and consistent bias assessment. Our method reduces human subjectivity and enhances fairness, scalability, and reproducibility in auditing GenAI-produced educational content.
title Automated Bias Assessment in AI-Generated Educational Content Using CEAT Framework
topic Computation and Language
Human-Computer Interaction
url https://arxiv.org/abs/2505.12718