Vision-Language Pre-training (VLP) models have achieved state-of-the-art performance in numerous cross-modal tasks. Since they are optimized to capture the statistical properties of intra- and inter-modality, there remains risk to learn social biases presented in the data as well. In this work, we (1) introduce a counterfactual-based bias measurement \emph{CounterBias} to quantify the social bias in VLP models by comparing the [MASK]ed prediction probabilities of factual and counterfactual samples; (2) construct a novel VL-Bias dataset including 24K image-text pairs for measuring gender bias in VLP models, from which we observed that significant gender bias is prevalent in VLP models; and (3) propose a VLP debiasing method \emph{FairVLP} to minimize the difference in the [MASK]ed prediction probabilities between factual and counterfactual image-text pairs for VLP debiasing. Although CounterBias and FairVLP focus on social bias, they are generalizable to serve as tools and provide new insights to probe and regularize more knowledge in VLP models.
翻译:在这项工作中,我们(1) 采用基于反事实的偏见衡量法,通过比较事实和反事实样本的[MASK]预测概率和反事实样本的[MASK]预测概率,量化VLP模型的社会偏见;(2) 建立一个新型VL-Bias数据集,包括24K的图像文本配对,用于测量VLP模型中的性别偏向,我们从这些模型中注意到,在VLP模型中普遍存在严重的性别偏向;(3) 提出VLP分化法,以尽量减少VLP在事实和反事实图像文本的[MASK]预测概率之间的差异;(2) 建立新型VL-Bias数据集,包括用于衡量VLP模型中性别偏向的24K图像文本配对,我们从中发现,在VLP模型中普遍存在严重的性别偏向;(3) 提议一个VLP分化法方法,以尽量减少VLP的预测概率和反事实图像文本对立的概率差异;虽然反对Bias和FairVLP对社会偏见的焦点是新的,但它们作为常规的探索工具,作为新的工具。