Are Hard Examples also Harder to Explain? A Study with Human and Model-Generated Explanations

31403

Knowledge Triples

Source Entity	Relationship	Target Entity
Few-Shot Prompting Method	enables	Large Pre-trained Language Models Material
Large Pre-trained Language Models Material	appliesTo	Explainable NLP OtherEntity
Sample Hardness OtherEntity	impacts	Generalizability Metric
Sample Hardness OtherEntity	impacts	Supportiveness Metric