arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2304.04874cs.CVcs.AIcs.LG
收藏

ImageCaptioner$^2$:用于图像描述偏差放大评估的图像描述器

ImageCaptioner$^2$: Image Captioner for Image Captioning Bias Amplification Assessment

King Abdullah University of Science and Technology (KAUST)(阿卜杜拉国王科技大学);National University of Singapore(新加坡国立大学);AWS AI, Amazon(亚马逊AWS人工智能)

Eslam Mohamed Bakr, Pengzhan Sun, Li Erran Li, Mohamed Elhoseiny

AI总结:

本文提出 ImageCaptioner$^2$ 偏差放大评估指标,通过纳入图像并采用基于提示的描述方式测量模型相对数据的偏差,并在多数据集、多属性和11种架构上验证其优于 LIC。

摘要:

众所周知,大多数预训练学习系统都受到偏差的困扰,这种偏差通常来自数据、模型或两者。测量和量化偏差及其来源是一项具有挑战性的任务,并且已在图像描述中得到广泛研究。尽管在这一方向上付出了巨大努力,但我们观察到,现有指标在纳入视觉信号方面缺乏一致性。在本文中,我们提出了一种新的偏差评估指标,称为 $ImageCaptioner^2$,用于图像描述。$ImageCaptioner^2$ 并不测量模型或数据中的绝对偏差,而是更关注模型相对于数据偏差所引入的偏差,这被称为偏差放大。现有方法仅根据生成的描述来评估图像描述算法,与之不同,$ImageCaptioner^2$ 在测量偏差时纳入了图像。此外,我们设计了一种将生成描述的偏差测量表述为基于提示的图像描述的形式,而不是使用语言分类器。最后,我们在三个不同数据集(即 MS-COCO caption 数据集、Artemis V1 和 Artemis V2)上,以及三个不同的受保护属性(即性别、种族和情绪)上,跨 11 种不同的图像描述架构应用了我们的 $ImageCaptioner^2$ 指标。因此,我们通过提出 AnonymousBench 验证了 $ImageCaptioner^2$ 指标的有效性,AnonymousBench 是一种用于偏差指标的新型人工评估范式。我们的指标在与人类一致性方面表现出相对于近期偏差指标 LIC 的显著优势,其中我们的指标和 LIC 的相关分数分别为 80% 和 54%。代码可在 https://eslambakr.github.io/imagecaptioner2.github.io/ 获取。

英文摘要:

Most pre-trained learning systems are known to suffer from bias, which typically emerges from the data, the model, or both. Measuring and quantifying bias and its sources is a challenging task and has been extensively studied in image captioning. Despite the significant effort in this direction, we observed that existing metrics lack consistency in the inclusion of the visual signal. In this paper, we introduce a new bias assessment metric, dubbed $ImageCaptioner^2$, for image captioning. Instead of measuring the absolute bias in the model or the data, $ImageCaptioner^2$ pay more attention to the bias introduced by the model w.r.t the data bias, termed bias amplification. Unlike the existing methods, which only evaluate the image captioning algorithms based on the generated captions only, $ImageCaptioner^2$ incorporates the image while measuring the bias. In addition, we design a formulation for measuring the bias of generated captions as prompt-based image captioning instead of using language classifiers. Finally, we apply our $ImageCaptioner^2$ metric across 11 different image captioning architectures on three different datasets, i.e., MS-COCO caption dataset, Artemis V1, and Artemis V2, and on three different protected attributes, i.e., gender, race, and emotions. Consequently, we verify the effectiveness of our $ImageCaptioner^2$ metric by proposing AnonymousBench, which is a novel human evaluation paradigm for bias metrics. Our metric shows significant superiority over the recent bias metric; LIC, in terms of human alignment, where the correlation scores are 80% and 54% for our metric and LIC, respectively. The code is available at https://eslambakr.github.io/imagecaptioner2.github.io/.

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。

↑