论文标题
减轻字幕系统中的性别偏见
Mitigating Gender Bias in Captioning Systems
论文作者
论文摘要
图像字幕通过来自网络的巨大支持图像收集取得了重大进展。但是，最近的研究指出，字幕数据集（例如可可）包含Web Corpora中发现的性别偏见。结果，学习模型可以在很大程度上依靠学习的先知和图像上下文来识别性别，从而导致不正确甚至是进攻性错误。为了鼓励模型学习正确的性别特征，我们重组可可数据集并介绍两个新的COCO-GB V1和V2数据集，其中火车和测试集具有不同的性别 - 封闭式分布。依靠上下文提示的模型将遭受反疾病测试数据的巨大性别预测错误。基准测试实验表明，大多数字幕模型都学习性别偏见，导致性别预测较高，尤其是对于女性而言。为了减轻不需要的偏见，我们提出了一种新的引导注意图像字幕模型（GAIC），该模型（GAIC）提供了视觉注意力的自我掩护，以鼓励该模型捕获正确的性别视觉证据。实验结果验证了盖克可以通过竞争性标题质量大大减少性别预测错误。我们的代码和设计的基准数据集可在https://github.com/datamllab/mitigating_gender_bias_in_in_captioning_system上获得。
Image captioning has made substantial progress with huge supporting image collections sourced from the web. However, recent studies have pointed out that captioning datasets, such as COCO, contain gender bias found in web corpora. As a result, learning models could heavily rely on the learned priors and image context for gender identification, leading to incorrect or even offensive errors. To encourage models to learn correct gender features, we reorganize the COCO dataset and present two new splits COCO-GB V1 and V2 datasets where the train and test sets have different gender-context joint distribution. Models relying on contextual cues will suffer from huge gender prediction errors on the anti-stereotypical test data. Benchmarking experiments reveal that most captioning models learn gender bias, leading to high gender prediction errors, especially for women. To alleviate the unwanted bias, we propose a new Guided Attention Image Captioning model (GAIC) which provides self-guidance on visual attention to encourage the model to capture correct gender visual evidence. Experimental results validate that GAIC can significantly reduce gender prediction errors with a competitive caption quality. Our codes and the designed benchmark datasets are available at https://github.com/datamllab/Mitigating_Gender_Bias_In_Captioning_System.
