TL;DR本文总结了解释性自然语言处理领域近年来集中收集人类注释的文本解释，将其分为三类 (highlights, free-text, and structured) 并对其注释方法进行整理，提出了现有采集方法的优点和缺点，给出了未来收集数据集的建议。
Abstract
explainable nlp (ExNLP) has increasingly focused on collecting human-annotated explanations. These explanations are used downstream in three ways: as data augmentation to improve performance on a →