Fetching the paper…

Pair the Dots: Jointly Examining Training History and Test Stimuli for Model Interpretability · Around