Fairness and Bias in Multimodal AI: A Survey

TL;DR

This survey addresses the gap in systematic study of fairness and bias in large multimodal models (LMMs) by conducting a structured literature review across LMMS and LLMS, and by cataloging 50 datasets and models with bias implications. It outlines a three-pronged mitigation framework (preprocessing, intrinsic, extrinsic) and highlights the underexplored preprocessing step 'preuse', along with multilingual bias and a spectrum of debiasing techniques. The authors synthesize audit/evaluation metrics, bias evaluation datasets, and an array of debiasing strategies, offering practical guidance for researchers and stakeholders while acknowledging limitations and the need for human-in-the-loop and societal impact considerations. Overall, the work provides a comprehensive roadmap for assessing, auditing, and reducing fairness and bias in multimodal and language AI, with clear implications for policy, ethics, and future research directions.

Abstract

The importance of addressing fairness and bias in artificial intelligence (AI) systems cannot be over-emphasized. Mainstream media has been awashed with news of incidents around stereotypes and other types of bias in many of these systems in recent years. In this survey, we fill a gap with regards to the relatively minimal study of fairness and bias in Large Multimodal Models (LMMs) compared to Large Language Models (LLMs), providing 50 examples of datasets and models related to both types of AI along with the challenges of bias affecting them. We discuss the less-mentioned category of mitigating bias, preprocessing (with particular attention on the first part of it, which we call preuse). The method is less-mentioned compared to the two well-known ones in the literature: intrinsic and extrinsic mitigation methods. We critically discuss the various ways researchers are addressing these challenges. Our method involved two slightly different search queries on two reputable search engines, Google Scholar and Web of Science (WoS), which revealed that for the queries 'Fairness and bias in Large Multimodal Models' and 'Fairness and bias in Large Language Models', 33,400 and 538,000 links are the initial results, respectively, for Scholar while 4 and 50 links are the initial results, respectively, for WoS. For reproducibility and verification, we provide links to the search results and the citations to all the final reviewed papers. We believe this work contributes to filling this gap and providing insight to researchers and other stakeholders on ways to address the challenges of fairness and bias in multimodal and language AI.