Text-to-image models, which can generate high-quality images based on textual
input, have recently enabled various content-creation tools. Despite
significantly affecting a wide range of downstream applications, the
distributions of these generated images are still not fully understood,
especially when it comes to the potential stereotypical attributes of different
genders. In this work, we propose a paradigm (Gender Presentation Differences)
that utilizes fine-grained self-presentation attributes to study how gender is
presented differently in text-to-image models. By probing gender indicators in
the input text (e.g., "a woman" or "a man"), we quantify the frequency
differences of presentation-centric attributes (e.g., "a shirt" and "a dress")
through human annotation and introduce a novel metric: GEP. Furthermore, we
propose an automatic method to estimate such differences. The automatic GEP
metric based on our approach yields a higher correlation with human annotations
than that based on existing CLIP scores, consistently across three
state-of-the-art text-to-image models. Finally, we demonstrate the
generalization ability of our metrics in the context of gender stereotypes
related to occupations.

本文提出了一种利用细粒度的自我表达属性研究文本到图像模型中不同性别呈现的范式（称为 Gender Presentation Differences 或 GPD），并通过人工注释量化性别指示符在输入文本中的频率差异并引入了一种新颖的指标：GEP，同时我们还提出了一种自动估计这种差异的方法。自动 GEP 指标基于我们的方法得出的相关性比基于现有 CLIP 得分的相关性更高，无论对于三个最先进的文本到图像模型如何一致，最后，在职业性别成见的背景下，我们展示了我们度量的普遍适用性。