将视觉分类表示为词的线性组合
Representing visual classification as a linear combination of words
AI总结：
提出一种利用视觉-语言模型将视觉分类任务表示为词线性组合的可解释性策略，通过预训练联合嵌入空间为词生成权重，实验表明该方法能有效对齐临床知识并帮助非专家执行专业医学任务。
摘要：
可解释性是深度学习中的一项长期挑战，尤其是在医疗保健等高风险领域。常见的可解释性方法通常会突出驱动 AI 模型做出决策的图像区域。然而，人类严重依赖语言来传达解释，不仅是“在哪里”，还包括“是什么”。此外，大多数可解释性方法侧重于解释 AI 的单个预测，而不是描述 AI 模型总体上使用的特征。后者对于模型和数据集审计将特别有用，随着 AI 越来越多地被用于新任务，它甚至可能促进知识的生成。在此，我们提出了一种可解释性策略，该策略使用视觉-语言模型来识别视觉分类任务中基于语言的描述符。通过利用图像和文本之间预训练的联合嵌入空间，我们的方法将一个新的分类任务估计为词的线性组合，从而为每个词生成一个权重，指示其与基于视觉的分类器的一致性。我们使用两项医学图像分类任务评估了该方法，我们发现，尽管缺乏特定领域的语言训练，但生成的描述符在很大程度上与临床知识相一致。然而，我们的方法也识别出了所用公开数据集中存在“捷径连接”的潜在可能。为了实现可解释性的功能性度量，我们进行了一项试点读者研究，发现 AI 识别出的词能够使非专家人类以非平凡的水平执行专业的医学任务。总而言之，我们的结果强调了使用多模态基础模型为视觉任务提供直观的、基于语言的解释的潜力。
英文摘要：
Explainability is a longstanding challenge in deep learning, especially in high-stakes domains like healthcare. Common explainability methods highlight image regions that drive an AI model's decision. Humans, however, heavily rely on language to convey explanations of not only "where" but "what". Additionally, most explainability approaches focus on explaining individual AI predictions, rather than describing the features used by an AI model in general. The latter would be especially useful for model and dataset auditing, and potentially even knowledge generation as AI is increasingly being used in novel tasks. Here, we present an explainability strategy that uses a vision-language model to identify language-based descriptors of a visual classification task. By leveraging a pre-trained joint embedding space between images and text, our approach estimates a new classification task as a linear combination of words, resulting in a weight for each word that indicates its alignment with the vision-based classifier. We assess our approach using two medical imaging classification tasks, where we find that the resulting descriptors largely align with clinical knowledge despite a lack of domain-specific language training. However, our approach also identifies the potential for 'shortcut connections' in the public datasets used. Towards a functional measure of explainability, we perform a pilot reader study where we find that the AI-identified words can enable non-expert humans to perform a specialized medical task at a non-trivial level. Altogether, our results emphasize the potential of using multimodal foundational models to deliver intuitive, language-based explanations of visual tasks.
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。
