arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.09014cs.SEcs.AI
收藏

生成模型如何绘制软件工程师?关于Stable Diffusion偏见的一项案例研究

How Do Generative Models Draw a Software Engineer? A Case Study on Stable Diffusion Bias

Department of Computer Science, University College London, London, UK(计算机科学系,伦敦大学学院,伦敦,英国);Department of Information Engineering, Computer Science and Mathematics, University of L'Aquila, L'Aquila, Italy(信息工程、计算机科学与数学系,拉奎拉大学,拉奎拉,意大利)

Tosin Fadahunsi, Giordano d'Aloisio, Antinisca Di Marco, Federica Sarro

AI总结:

本研究通过实证评估Stable Diffusion三个版本在软件工程任务中的图像生成,发现模型在性别和种族上存在显著偏见,凸显了采用生成模型需警惕偏见风险。

摘要:

生成模型如今被广泛用于生成图形内容,用于多种目的,例如网页、艺术、广告。然而,已有研究表明,这些模型生成的图像可能强化特定情境中已存在的社会偏见。在本文中,我们专注于理解当生成与各种软件工程任务相关的图像时,情况是否如此。事实上,软件工程(SE)社区并非不受性别和种族差异的影响,而这些差异可能因使用这些模型而被放大。因此,如果在无意识的情况下使用,人工生成的图像可能在SE领域强化这些偏见。具体而言,我们对Stable Diffusion(SD)模型(一个非常流行的开源文本到图像模型)的三个版本——SD 2、SD XL和SD 3——在SE任务中暴露的性别和种族偏见进行了广泛的实证评估。我们通过向每个模型提供两组描述不同软件相关任务的提示词,获得了6,720张图像:一组包含“软件工程师”关键词,另一组不包含任何关于执行任务人员的说明。接下来,我们评估了生成图像中的性别和种族差异。结果表明,在描绘软件工程师时,所有模型都显著偏向于男性形象。相反,虽然SD 2和SD XL强烈偏向于白人形象,但SD 3则略微偏向于亚洲人形象。然而,无论使用何种提示风格,所有模型都显著低估了黑人和阿拉伯人形象。我们的分析结果凸显了采用这些模型为SE任务生成内容的严重担忧,并为未来在该情境下进行偏见缓解研究开辟了领域。

英文摘要:

Generative models are nowadays widely used to generate graphical content used for multiple purposes, e.g. web, art, advertisement. However, it has been shown that the images generated by these models could reinforce societal biases already existing in specific contexts. In this paper, we focus on understanding if this is the case when one generates images related to various software engineering tasks. In fact, the Software Engineering (SE) community is not immune from gender and ethnicity disparities, which could be amplified by the use of these models. Hence, if used without consciousness, artificially generated images could reinforce these biases in the SE domain. Specifically, we perform an extensive empirical evaluation of the gender and ethnicity bias exposed by three versions of the Stable Diffusion (SD) model (a very popular open-source text-to-image model) - SD 2, SD XL, and SD 3 - towards SE tasks. We obtain 6,720 images by feeding each model with two sets of prompts describing different software-related tasks: one set includes the Software Engineer keyword, and one set does not include any specification of the person performing the task. Next, we evaluate the gender and ethnicity disparities in the generated images. Results show how all models are significantly biased towards male figures when representing software engineers. On the contrary, while SD 2 and SD XL are strongly biased towards White figures, SD 3 is slightly more biased towards Asian figures. Nevertheless, all models significantly under-represent Black and Arab figures, regardless of the prompt style used. The results of our analysis highlight severe concerns about adopting those models to generate content for SE tasks and open the field for future research on bias mitigation in this context.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑