Keywords: NLP, explainability, explainable NLP, explainable AI, XAI
TL;DR: We identify datasets with 3 classes of textual explanations, organize the literature on annotating each type, identify strengths/shortcomings of existing collection methodologies, and give recommendations for collecting explanations in the future.
Abstract: Explainable Natural Language Processing (ExNLP) has increasingly focused on collecting human-annotated textual explanations. These explanations are used downstream in three ways: as data augmentation to improve performance on a predictive task, as supervision to train models to produce explanations for their predictions, and as a ground-truth to evaluate model-generated explanations. In this review, we identify 65 datasets with three predominant classes of textual explanations (highlights, free-text, and structured), organize the literature on annotating each type, identify strengths and shortcomings of existing collection methodologies, and give recommendations for collecting ExNLP datasets in the future.
Supplementary Material: zip
Contribution Process Agreement: Yes
Dataset Url: https://exnlpdatasets.github.io/
Dataset Embargo: N/A
Author Statement: Yes