Operationalizing Fairness in Text-to-Image Models: A Survey of Bias, Fairness Audits and Mitigation Strategies

TL;DR

This survey reviews societal biases in text-to-image models, categorizes existing fairness approaches, and proposes a new framework for operational fairness to improve accountability in generative AI.

Contribution

It provides a systematic taxonomy of bias and fairness in T2I models, critically assesses current mitigation strategies, and introduces a novel operational fairness framework.

Findings

Identifies conceptual ambiguities in fairness terminology.

Classifies bias types and fairness notions in T2I models.

Proposes a target-based testing framework for fairness.

Abstract

Text-to-Image (T2I) generation models have been widely adopted across various industries, yet are criticized for frequently exhibiting societal stereotypes. While a growing body of research has emerged to evaluate and mitigate these biases, the field at present contends with conceptual ambiguity, for example terms like "bias" and "fairness" are not always clearly distinguished and often lack clear operational definitions. This paper provides a comprehensive systematic review of T2I fairness literature, organizing existing work into a taxonomy of bias types and fairness notions. We critically assess the gap between "target fairness" (normative ideals in T2I outputs) and "threshold fairness" (normative standards with actionable decision rules). Furthermore, we survey the landscape of mitigation strategies, ranging from prompt engineering to diffusion process manipulation. We conclude by proposing a new framework for operationalizing fairness that moves beyond descriptive metrics towards rigorous, target-based testing, offering an approach for more accountable generative AI development.