Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies
<p dir="ltr">Despite their continuous advancements, text-to-image (TTI) models often reflect and reinforce cultural biases, perpetuating stereotypes often inherent in their training data. This systematic review critically examines cultural bias in text-to-image (TTI) models, addressi...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , |
| Published: |
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1864513531587592192 |
|---|---|
| author | Wala Elsharif (22828082) |
| author2 | Mahmood Alzubaidi (15740693) Marco Agus (8032898) |
| author2_role | author author |
| author_facet | Wala Elsharif (22828082) Mahmood Alzubaidi (15740693) Marco Agus (8032898) |
| author_role | author |
| dc.creator.none.fl_str_mv | Wala Elsharif (22828082) Mahmood Alzubaidi (15740693) Marco Agus (8032898) |
| dc.date.none.fl_str_mv | 2025-07-18T12:00:00Z |
| dc.identifier.none.fl_str_mv | 10.1109/access.2025.3585745 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/journal_contribution/Cultural_Bias_in_Text-to-Image_Models_A_Systematic_Review_of_Bias_Identification_Evaluation_and_Mitigation_Strategies/30860168 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Information and computing sciences Artificial intelligence Computer vision and multimedia computation Philosophy and religious studies Applied ethics AI ethics AI fairness bias evaluation bias mitigation CLIP cultural bias generative AI gender bias prompt engineering racial bias responsible AI text-to-image models |
| dc.title.none.fl_str_mv | Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies |
| dc.type.none.fl_str_mv | Text Journal contribution info:eu-repo/semantics/publishedVersion text contribution to journal |
| description | <p dir="ltr">Despite their continuous advancements, text-to-image (TTI) models often reflect and reinforce cultural biases, perpetuating stereotypes often inherent in their training data. This systematic review critically examines cultural bias in text-to-image (TTI) models, addressing gaps in existing research by analyzing its manifestations, evaluation methods, and mitigation strategies—both directly and through the lens of intersectionality with other bias dimensions. A comprehensive literature review was conducted across multiple major databases, following a rigorously structured search strategy, resulting in the selection of 58 studies spanning bias analysis, evaluation frameworks, and mitigation techniques. Thematic findings highlight that gender bias was the most extensively studied, appearing in 53 studies (91%), followed by racial/ethnic bias (42 studies) and other social biases (41 studies). Furthermore, the review explores how these biases intersect and compound in AI-generated imagery, shaping and reinforcing cultural bias. Our findings reveal the following key aspects: 1) the lack of standardization and scalability in bias evaluation, 2) the lack of a fully effective mitigation strategy, 3) contributed TTI benchmarks favoring Western-centric perspectives. We finally propose future directions to improve fairness and representation in TTI models.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2025.3585745" target="_blank">https://dx.doi.org/10.1109/access.2025.3585745</a></p> |
| eu_rights_str_mv | openAccess |
| id | Manara2_996bbff98b4f38d2cd0ba4c8e63b1a67 |
| identifier_str_mv | 10.1109/access.2025.3585745 |
| network_acronym_str | Manara2 |
| network_name_str | Manara2 |
| oai_identifier_str | oai:figshare.com:article/30860168 |
| publishDate | 2025 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation StrategiesWala Elsharif (22828082)Mahmood Alzubaidi (15740693)Marco Agus (8032898)Information and computing sciencesArtificial intelligenceComputer vision and multimedia computationPhilosophy and religious studiesApplied ethicsAI ethicsAI fairnessbias evaluationbias mitigationCLIPcultural biasgenerative AIgender biasprompt engineeringracial biasresponsible AItext-to-image models<p dir="ltr">Despite their continuous advancements, text-to-image (TTI) models often reflect and reinforce cultural biases, perpetuating stereotypes often inherent in their training data. This systematic review critically examines cultural bias in text-to-image (TTI) models, addressing gaps in existing research by analyzing its manifestations, evaluation methods, and mitigation strategies—both directly and through the lens of intersectionality with other bias dimensions. A comprehensive literature review was conducted across multiple major databases, following a rigorously structured search strategy, resulting in the selection of 58 studies spanning bias analysis, evaluation frameworks, and mitigation techniques. Thematic findings highlight that gender bias was the most extensively studied, appearing in 53 studies (91%), followed by racial/ethnic bias (42 studies) and other social biases (41 studies). Furthermore, the review explores how these biases intersect and compound in AI-generated imagery, shaping and reinforcing cultural bias. Our findings reveal the following key aspects: 1) the lack of standardization and scalability in bias evaluation, 2) the lack of a fully effective mitigation strategy, 3) contributed TTI benchmarks favoring Western-centric perspectives. We finally propose future directions to improve fairness and representation in TTI models.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2025.3585745" target="_blank">https://dx.doi.org/10.1109/access.2025.3585745</a></p>2025-07-18T12:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/access.2025.3585745https://figshare.com/articles/journal_contribution/Cultural_Bias_in_Text-to-Image_Models_A_Systematic_Review_of_Bias_Identification_Evaluation_and_Mitigation_Strategies/30860168CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/308601682025-07-18T12:00:00Z |
| spellingShingle | Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies Wala Elsharif (22828082) Information and computing sciences Artificial intelligence Computer vision and multimedia computation Philosophy and religious studies Applied ethics AI ethics AI fairness bias evaluation bias mitigation CLIP cultural bias generative AI gender bias prompt engineering racial bias responsible AI text-to-image models |
| status_str | publishedVersion |
| title | Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies |
| title_full | Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies |
| title_fullStr | Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies |
| title_full_unstemmed | Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies |
| title_short | Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies |
| title_sort | Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies |
| topic | Information and computing sciences Artificial intelligence Computer vision and multimedia computation Philosophy and religious studies Applied ethics AI ethics AI fairness bias evaluation bias mitigation CLIP cultural bias generative AI gender bias prompt engineering racial bias responsible AI text-to-image models |