VLBiasBench: A Comprehensive Benchmark for Evaluating Bias in Large Vision-Language Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Sibo, Cao, Xiangkui, Zhang, Jie, Yuan, Zheng, Shan, Shiguang, Chen, Xilin, Gao, Wen
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911567979741184
author Wang, Sibo
Cao, Xiangkui
Zhang, Jie
Yuan, Zheng
Shan, Shiguang
Chen, Xilin
Gao, Wen
author_facet Wang, Sibo
Cao, Xiangkui
Zhang, Jie
Yuan, Zheng
Shan, Shiguang
Chen, Xilin
Gao, Wen
contents The emergence of Large Vision-Language Models (LVLMs) marks significant strides towards achieving general artificial intelligence. However, these advancements are accompanied by concerns about biased outputs, a challenge that has yet to be thoroughly explored. Existing benchmarks are not sufficiently comprehensive in evaluating biases due to their limited data scale, single questioning format and narrow sources of bias. To address this problem, we introduce VLBiasBench, a comprehensive benchmark designed to evaluate biases in LVLMs. VLBiasBench, features a dataset that covers nine distinct categories of social biases, including age, disability status, gender, nationality, physical appearance, race, religion, profession, social economic status, as well as two intersectional bias categories: race x gender and race x social economic status. To build a large-scale dataset, we use Stable Diffusion XL model to generate 46,848 high-quality images, which are combined with various questions to creat 128,342 samples. These questions are divided into open-ended and close-ended types, ensuring thorough consideration of bias sources and a comprehensive evaluation of LVLM biases from multiple perspectives. We conduct extensive evaluations on 15 open-source models as well as two advanced closed-source models, yielding new insights into the biases present in these models. Our benchmark is available at https://github.com/Xiangkui-Cao/VLBiasBench.
format Preprint
id arxiv_https___arxiv_org_abs_2406_14194
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle VLBiasBench: A Comprehensive Benchmark for Evaluating Bias in Large Vision-Language Model
Wang, Sibo
Cao, Xiangkui
Zhang, Jie
Yuan, Zheng
Shan, Shiguang
Chen, Xilin
Gao, Wen
Computer Vision and Pattern Recognition
Artificial Intelligence
The emergence of Large Vision-Language Models (LVLMs) marks significant strides towards achieving general artificial intelligence. However, these advancements are accompanied by concerns about biased outputs, a challenge that has yet to be thoroughly explored. Existing benchmarks are not sufficiently comprehensive in evaluating biases due to their limited data scale, single questioning format and narrow sources of bias. To address this problem, we introduce VLBiasBench, a comprehensive benchmark designed to evaluate biases in LVLMs. VLBiasBench, features a dataset that covers nine distinct categories of social biases, including age, disability status, gender, nationality, physical appearance, race, religion, profession, social economic status, as well as two intersectional bias categories: race x gender and race x social economic status. To build a large-scale dataset, we use Stable Diffusion XL model to generate 46,848 high-quality images, which are combined with various questions to creat 128,342 samples. These questions are divided into open-ended and close-ended types, ensuring thorough consideration of bias sources and a comprehensive evaluation of LVLM biases from multiple perspectives. We conduct extensive evaluations on 15 open-source models as well as two advanced closed-source models, yielding new insights into the biases present in these models. Our benchmark is available at https://github.com/Xiangkui-Cao/VLBiasBench.
title VLBiasBench: A Comprehensive Benchmark for Evaluating Bias in Large Vision-Language Model
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2406.14194