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

TL;DR

VLBiasBench is a comprehensive benchmark dataset designed to evaluate biases in large vision-language models across multiple social bias categories using generated images and diverse questions.

Contribution

It introduces a large-scale, multi-category bias evaluation benchmark for LVLMs, addressing limitations of previous benchmarks in scope and diversity.

Findings

Extensive bias evaluations on 17 LVLMs reveal prevalent social biases.

The benchmark uncovers differences in bias levels between open-source and closed-source models.

VLBiasBench provides a new resource for systematic bias assessment in LVLMs.

Abstract

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.