Ablating Concepts in Text-to-Image Diffusion Models

doi:doi:10.57702/zh0aiuag

Ablating Concepts in Text-to-Image Diffusion Models

Large-scale text-to-image diffusion models can gener-ate high-fidelity images with powerful compositional ability. However, these models are typically trained on an enormous amount of Internet data, often containing copyrighted material, licensed images, and personal photos.

BibTex:

@dataset{Nupur_Kumari_and_Bingliang_Zhang_and_Sheng-Yu_Wang_and_Eli_Shechtman_and_Richard_Zhang_and_Jun-Yan_Zhu_2024,
    abstract = {Large-scale text-to-image diffusion models can gener-ate high-fidelity images with powerful compositional ability. However, these models are typically trained on an enormous amount of Internet data, often containing copyrighted material, licensed images, and personal photos.},
    author = {Nupur Kumari and Bingliang Zhang and Sheng-Yu Wang and Eli Shechtman and Richard Zhang and Jun-Yan Zhu},
    doi = {10.57702/zh0aiuag},
    institution = {No Organization},
    keyword = {'Concept Ablation', 'Text-to-Image Synthesis', 'copyrighted material', 'image generation', 'licensed images', 'personal photos', 'text-to-image diffusion models'},
    month = {dec},
    publisher = {TIB},
    title = {Ablating Concepts in Text-to-Image Diffusion Models},
    url = {https://service.tib.eu/ldmservice/dataset/ablating-concepts-in-text-to-image-diffusion-models},
    year = {2024}
}