Explainable Cultural Bias Auditing Framework for Text-to-Image Generative Models Across Multilingual Contexts
Keywords:
cultural bias, text-to-image generation, explainable AI, multilingual auditing, fairness infrastructure, diffusion models, governanceAbstract
The proliferation of text-to-image generative models has introduced profound challenges in understanding and mitigating cultural biases that manifest across languages and visual representations. Current bias auditing remains fragmented, often focusing on isolated demographic dimensions in a single language, without systematically explaining why culturally skewed outputs emerge or how they propagate across multilingual prompts. This paper proposes the Explainable Cultural Bias Auditing Framework, an integrative system architecture designed to detect, interpret, and govern cultural bias in text-to-image pipelines spanning multiple languages. The framework combines multilingual prompt curation, cross-modal concept probing, gradient-based and attention-driven explainability layers, and governance interfaces that connect internal model behavior with external fairness standards. We examine the structural trade-offs between audit depth, computational scalability, and real-time deployment; analyze the interdependency between data provenance, model architecture, and cultural representational gaps; and situate the framework within broader policy landscapes, including the EU AI Act and the NIST AI Risk Management Framework. Through a systems-level discussion, we highlight how infrastructure choices, sustainability constraints, and explainability granularity collectively shape the robustness of cultural bias detection. Rather than proposing a single algorithm, the paper articulates a principled architectural paradigm that integrates fairness-aware infrastructure, multilingual semantic alignment, and interpretability-first design, providing a scaffold for future auditing systems in generative AI ecosystems.
References
1. Buolamwini, J., & Gebru, T. (2018). Gender shades: Intersectional accuracy disparities in commercial gender classification. Proceedings of the 1st Conference on Fairness, Accountability and Transparency, 77-91.
2. Mitchell, M., Wu, S., Zaldivar, A., Barnes, P., Vasserman, L., Hutchinson, B., ... & Gebru, T. (2019). Model cards for model reporting. In Proceedings of the Conference on Fairness, Accountability, and Transparency (pp. 220-229). ACM.
3. Bender, E. M., & Friedman, B. (2018). Data statements for natural language processing: Toward mitigating system bias and enabling better science. Transactions of the Association for Computational Linguistics, 6, 587-604.
4. Bolukbasi, T., Chang, K. W., Zou, J. Y., Saligrama, V., & Kalai, A. T. (2016). Man is to computer programmer as woman is to homemaker? Debiasing word embeddings. In Advances in Neural Information Processing Systems (pp. 4349-4357).
5. Zhao, J., Wang, T., Yatskar, M., Ordonez, V., & Chang, K. W. (2017). Men also like shopping: Reducing gender bias amplification using corpus-level constraints. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (pp. 2979-2989). ACL.
6. Sheng, E., Chang, K. W., Natarajan, P., & Peng, N. (2019). The woman worked as a babysitter: On biases in language generation. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (pp. 3405-3410). ACL.
7. Nadeem, M., Bethke, A., & Reddy, S. (2021). StereoSet: Measuring stereotypical bias in pretrained language models. In Proceedings of the Association for Computational Linguistics (pp. 5356-5371). ACL.
8. Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2022). High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 10684-10695). IEEE.
9. Ramesh, A., Pavlov, M., Goh, G., Gray, S., Voss, C., Radford, A., ... & Sutskever, I. (2021). Zero-shot text-to-image generation. In International Conference on Machine Learning (pp. 8821-8831). PMLR.
10. Saharia, C., Chan, W., Saxena, S., Li, L., Whang, J., Denton, E., ... & Norouzi, M. (2022). Photorealistic text-to-image diffusion models with deep language understanding. In Advances in Neural Information Processing Systems.
11. Nichol, A., Dhariwal, P., Ramesh, A., Shyam, P., Mishkin, P., McGrew, B., ... & Chen, M. (2021). GLIDE: Towards photorealistic image generation and editing with text-guided diffusion models. In International Conference on Machine Learning (pp. 16784-16804). PMLR.
12. Luccioni, A. S., Akkaya, C., Jernite, Y., & others. (2023). Stable Bias: Analyzing societal representations in diffusion models. arXiv preprint arXiv:2303.11408.
13. Bianchi, F., Kalluri, P., Durmus, E., Ladhak, F., Cheng, M., Nozza, D., ... & Caliskan, A. (2023). Easily accessible text-to-image generation amplifies demographic stereotypes. In Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency (pp. 1493-1504). ACM.
14. Cho, J., Zala, A., Bansal, M., Shah, D., & Darrell, T. (2023). DALL-Eval: Probing the reasoning skills and social biases of text-to-image generation models. In Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops. IEEE.
15. Wang, Z., et al. (2023). How well can text-to-image generative models understand ethical natural language interventions? In Proceedings of the Conference on Empirical Methods in Natural Language Processing. ACL.
16. Carlsson, F., Eisen, P., Rekath, N., & Sahlgren, M. (2022). Cross-lingual and multilingual CLIP. In Proceedings of the Language Resources and Evaluation Conference (pp. 6848-6854). ELRA.
17. C. Shi, S. Li, S. Guo, S. Xie, W. Wu, J. Dou, C. Wu, C. Xiao, C. Wang, Z. Cheng, et al. (2025)Where culture fades: revealing the cultural gap in text-to-image generation.arXiv preprint arXiv:2511.17282.
18. Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). "Why should I trust you?": Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135-1144). ACM.
19. Lundberg, S. M., & Lee, S. I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems (pp. 4765-4774).
20. Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE International Conference on Computer Vision (pp. 618-626). IEEE.
21. Tang, R., Pandey, R., Qi, Z., & Saunshi, N. (2023). What the DAAM: Interpreting Stable Diffusion Using Cross Attention. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). ACL.
22. European Commission. (2021). Proposal for a Regulation of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act). COM(2021) 206 final.
23. National Institute of Standards and Technology. (2023). Artificial intelligence risk management framework (AI RMF 1.0). NIST.
24. Strubell, E., Ganesh, A., & McCallum, A. (2019). Energy and policy considerations for deep learning in NLP. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (pp. 3645-3650). ACL.
25. Barocas, S., Hardt, M., & Narayanan, A. (2019). Fairness and machine learning. fairmlbook.org.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Advanced Artificial Intelligence Research
This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
