The official repository for paper " Multi-Modal Bias: Introducing a Framework for Stereotypical Bias Assessment beyond Gender and Race in Vision–Language Models" by Sepehr Janghorbani, Gerard de Melo, published at the European Chapter of the Association for Computational Linguistics (EACL) 2023.
Image and textual dataset are found under data. To extract images Run the following comand:
Unzip <category.zip>where category is the minority subgroup under study.
Distributed under the MIT License. See LICENSE.txt for more information.
If you use our any part of our dataset or code, please cite our paper.
@article{janghorbani2023multimodal,
title={MultiModal Bias: Introducing a Framework for Stereotypical Bias Assessment beyond Gender and Race in Vision Language Models},
author={Janghorbani, Sepehr and de Melo, Gerard},
journal={arXiv preprint arXiv:2303.12734},
year={2023}
}