ViG-Bias：视觉扎根的偏差发现与缓解
ViG-Bias: Visually Grounded Bias Discovery and Mitigation
AI总结：
本文提出 ViG-Bias，通过引入 GradCAM 等视觉解释增强现有偏差发现与缓解框架，并在 CelebA、Waterbirds 和 NICO++ 上验证其可提升 DOMINO、FACTS 和 Bias-to-Text 的效果。
摘要：
机器学习模型在关键决策过程中的广泛应用，凸显了偏差发现与缓解策略的必要性。识别偏差系统背后的原因并不简单，因为在许多情况下，这些原因与难以发现的隐性虚假相关性有关。标准方法依赖于偏差审计，其做法是分析模型在预定义数据样本子群中的性能；这些子群通常由共同属性刻画，例如涉及人物时的性别或种族，或定义语义一致图像组的其他特定属性。然而，人们并不总能事先知道定义视觉识别系统失败模式的具体属性。近期方法提出利用大型视觉语言模型来发现这些群体，这类模型能够提取跨模态嵌入并生成文本描述，以刻画某个模型表现不佳的子群。在本研究中，我们认为纳入视觉解释（例如通过 GradCAM 或其他方法生成的热力图）可以提升此类偏差发现与缓解框架的性能。为此，我们提出了视觉扎根偏差发现与缓解（Visually Grounded Bias Discovery and Mitigation，ViG-Bias），这是一种简单而有效的技术，可集成到多种现有框架中，同时改进发现与缓解性能。我们的综合评估表明，纳入视觉解释可在多个具有挑战性的数据集（包括 CelebA、Waterbirds 和 NICO++）上增强 DOMINO、FACTS 和 Bias-to-Text 等现有技术。
英文摘要：
The proliferation of machine learning models in critical decision making processes has underscored the need for bias discovery and mitigation strategies. Identifying the reasons behind a biased system is not straightforward, since in many occasions they are associated with hidden spurious correlations which are not easy to spot. Standard approaches rely on bias audits performed by analyzing model performance in pre-defined subgroups of data samples, usually characterized by common attributes like gender or ethnicity when it comes to people, or other specific attributes defining semantically coherent groups of images. However, it is not always possible to know a-priori the specific attributes defining the failure modes of visual recognition systems. Recent approaches propose to discover these groups by leveraging large vision language models, which enable the extraction of cross-modal embeddings and the generation of textual descriptions to characterize the subgroups where a certain model is underperforming. In this work, we argue that incorporating visual explanations (e.g. heatmaps generated via GradCAM or other approaches) can boost the performance of such bias discovery and mitigation frameworks. To this end, we introduce Visually Grounded Bias Discovery and Mitigation (ViG-Bias), a simple yet effective technique which can be integrated to a variety of existing frameworks to improve both, discovery and mitigation performance. Our comprehensive evaluation shows that incorporating visual explanations enhances existing techniques like DOMINO, FACTS and Bias-to-Text, across several challenging datasets, including CelebA, Waterbirds, and NICO++.
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。
