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

Megan Smith, Venkatesh Thirugnana Sambandham, Florian Richter, Laura Crompton, Matthias Uhl, Torsten Schön
Proceedings of The Algorithmic Fairness Across Alignment Procedures and Agentic Systems, PMLR 344:72-92, 2026.

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v344-smith26a, title = {Operationalizing Fairness in Text-to-Image Models: A Survey of Bias, Fairness Audits and Mitigation Strategies}, author = {Smith, Megan and Thirugnana Sambandham, Venkatesh and Richter, Florian and Crompton, Laura and Uhl, Matthias and Sch\"{o}n, Torsten}, booktitle = {Proceedings of The Algorithmic Fairness Across Alignment Procedures and Agentic Systems}, pages = {72--92}, year = {2026}, editor = {Tang, Zeyu and Ganesh, Prakhar and Dieng, Awa and Rateike, Miriam and Watson-Daniels, Jamelle and Farnadi, Golnoosh}, volume = {344}, series = {Proceedings of Machine Learning Research}, month = {26 Apr}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v344/main/assets/smith26a/smith26a.pdf}, url = {https://proceedings.mlr.press/v344/smith26a.html}, 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.} }
Endnote
%0 Conference Paper %T Operationalizing Fairness in Text-to-Image Models: A Survey of Bias, Fairness Audits and Mitigation Strategies %A Megan Smith %A Venkatesh Thirugnana Sambandham %A Florian Richter %A Laura Crompton %A Matthias Uhl %A Torsten Schön %B Proceedings of The Algorithmic Fairness Across Alignment Procedures and Agentic Systems %C Proceedings of Machine Learning Research %D 2026 %E Zeyu Tang %E Prakhar Ganesh %E Awa Dieng %E Miriam Rateike %E Jamelle Watson-Daniels %E Golnoosh Farnadi %F pmlr-v344-smith26a %I PMLR %P 72--92 %U https://proceedings.mlr.press/v344/smith26a.html %V 344 %X 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.
APA
Smith, M., Thirugnana Sambandham, V., Richter, F., Crompton, L., Uhl, M. & Schön, T.. (2026). Operationalizing Fairness in Text-to-Image Models: A Survey of Bias, Fairness Audits and Mitigation Strategies. Proceedings of The Algorithmic Fairness Across Alignment Procedures and Agentic Systems, in Proceedings of Machine Learning Research 344:72-92 Available from https://proceedings.mlr.press/v344/smith26a.html.

Related Material