TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2023
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917724079259648 |
|---|---|
| author | Chinchure, Aditya Shukla, Pushkar Bhatt, Gaurav Salij, Kiri Hosanagar, Kartik Sigal, Leonid Turk, Matthew |
| author_facet | Chinchure, Aditya Shukla, Pushkar Bhatt, Gaurav Salij, Kiri Hosanagar, Kartik Sigal, Leonid Turk, Matthew |
| contents | Text-to-Image (TTI) generative models have shown great progress in the past few years in terms of their ability to generate complex and high-quality imagery. At the same time, these models have been shown to suffer from harmful biases, including exaggerated societal biases (e.g., gender, ethnicity), as well as incidental correlations that limit such a model's ability to generate more diverse imagery. In this paper, we propose a general approach to study and quantify a broad spectrum of biases, for any TTI model and for any prompt, using counterfactual reasoning. Unlike other works that evaluate generated images on a predefined set of bias axes, our approach automatically identifies potential biases that might be relevant to the given prompt, and measures those biases. In addition, we complement quantitative scores with post-hoc explanations in terms of semantic concepts in the images generated. We show that our method is uniquely capable of explaining complex multi-dimensional biases through semantic concepts, as well as the intersectionality between different biases for any given prompt. We perform extensive user studies to illustrate that the results of our method and analysis are consistent with human judgements. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_01261 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Models Chinchure, Aditya Shukla, Pushkar Bhatt, Gaurav Salij, Kiri Hosanagar, Kartik Sigal, Leonid Turk, Matthew Computer Vision and Pattern Recognition Computers and Society Text-to-Image (TTI) generative models have shown great progress in the past few years in terms of their ability to generate complex and high-quality imagery. At the same time, these models have been shown to suffer from harmful biases, including exaggerated societal biases (e.g., gender, ethnicity), as well as incidental correlations that limit such a model's ability to generate more diverse imagery. In this paper, we propose a general approach to study and quantify a broad spectrum of biases, for any TTI model and for any prompt, using counterfactual reasoning. Unlike other works that evaluate generated images on a predefined set of bias axes, our approach automatically identifies potential biases that might be relevant to the given prompt, and measures those biases. In addition, we complement quantitative scores with post-hoc explanations in terms of semantic concepts in the images generated. We show that our method is uniquely capable of explaining complex multi-dimensional biases through semantic concepts, as well as the intersectionality between different biases for any given prompt. We perform extensive user studies to illustrate that the results of our method and analysis are consistent with human judgements. |
| title | TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Models |
| topic | Computer Vision and Pattern Recognition Computers and Society |
| url | https://arxiv.org/abs/2312.01261 |