Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2502.08914 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912448297041920 |
|---|---|
| author | Bayramli, Zahra Suleymanzade, Ayhan An, Na Min Ahmad, Huzama Kim, Eunsu Park, Junyeong Thorne, James Oh, Alice |
| author_facet | Bayramli, Zahra Suleymanzade, Ayhan An, Na Min Ahmad, Huzama Kim, Eunsu Park, Junyeong Thorne, James Oh, Alice |
| contents | Text-to-image diffusion models have recently enabled the creation of visually compelling, detailed images from textual prompts. However, their ability to accurately represent various cultural nuances remains an open question. In our work, we introduce CultDiff benchmark, evaluating state-of-the-art diffusion models whether they can generate culturally specific images spanning ten countries. We show that these models often fail to generate cultural artifacts in architecture, clothing, and food, especially for underrepresented country regions, by conducting a fine-grained analysis of different similarity aspects, revealing significant disparities in cultural relevance, description fidelity, and realism compared to real-world reference images. With the collected human evaluations, we develop a neural-based image-image similarity metric, namely, CultDiff-S, to predict human judgment on real and generated images with cultural artifacts. Our work highlights the need for more inclusive generative AI systems and equitable dataset representation over a wide range of cultures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_08914 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Diffusion Models Through a Global Lens: Are They Culturally Inclusive? Bayramli, Zahra Suleymanzade, Ayhan An, Na Min Ahmad, Huzama Kim, Eunsu Park, Junyeong Thorne, James Oh, Alice Computer Vision and Pattern Recognition Artificial Intelligence Text-to-image diffusion models have recently enabled the creation of visually compelling, detailed images from textual prompts. However, their ability to accurately represent various cultural nuances remains an open question. In our work, we introduce CultDiff benchmark, evaluating state-of-the-art diffusion models whether they can generate culturally specific images spanning ten countries. We show that these models often fail to generate cultural artifacts in architecture, clothing, and food, especially for underrepresented country regions, by conducting a fine-grained analysis of different similarity aspects, revealing significant disparities in cultural relevance, description fidelity, and realism compared to real-world reference images. With the collected human evaluations, we develop a neural-based image-image similarity metric, namely, CultDiff-S, to predict human judgment on real and generated images with cultural artifacts. Our work highlights the need for more inclusive generative AI systems and equitable dataset representation over a wide range of cultures. |
| title | Diffusion Models Through a Global Lens: Are They Culturally Inclusive? |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2502.08914 |