SC-ML：用于去偏视觉问答的自监督反事实度量学习
SC-ML: Self-supervised Counterfactual Metric Learning for Debiased Visual Question Answering
AI总结：
针对VQA中的语言偏置问题，提出自监督反事实度量学习方法SC-ML，通过自适应选择问题相关视觉特征并融入反事实训练方案，在VQA-CP数据集上有效提升了模型去偏性能与鲁棒性。
摘要：
视觉问答（VQA）是一项关键的多模态任务，智能体必须根据视觉线索回答问题。不幸的是，语言偏置是VQA中的一个常见问题，指模型仅通过与问题关联而忽略视觉内容来生成答案，导致偏倚结果。我们通过提出自监督反事实度量学习（SC-ML）方法来解决语言偏置问题，以更好地聚焦图像特征。SC-ML能自适应地选择与问题相关的视觉特征来回答问题，减少与问题无关的视觉特征对推断答案的负面影响。此外，与问题无关的视觉特征可无缝融入反事实训练方案中，进一步增强鲁棒性。大量实验证明了我们方法的有效性，在VQA-CP数据集上取得了提升的结果。我们的代码将公开提供。
英文摘要：
Visual question answering (VQA) is a critical multimodal task in which an agent must answer questions according to the visual cue. Unfortunately, language bias is a common problem in VQA, which refers to the model generating answers only by associating with the questions while ignoring the visual content, resulting in biased results. We tackle the language bias problem by proposing a self-supervised counterfactual metric learning (SC-ML) method to focus the image features better. SC-ML can adaptively select the question-relevant visual features to answer the question, reducing the negative influence of question-irrelevant visual features on inferring answers. In addition, question-irrelevant visual features can be seamlessly incorporated into counterfactual training schemes to further boost robustness. Extensive experiments have proved the effectiveness of our method with improved results on the VQA-CP dataset. Our code will be made publicly available.
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。
