Let the CAT out of the bag: Contrastive Attributed explanations for Text

TL;DR

This paper introduces CAT, a novel method for generating contrastive explanations for text models by leveraging attribute classifiers, resulting in more meaningful, fluent, and minimally perturbed contrastive texts that outperform existing methods.

Contribution

The paper proposes a new approach using attribute classifiers and minimal perturbations to produce semantically rich contrastive explanations for text models.

Findings

Outperforms state-of-the-art methods on four datasets

Produces more semantically meaningful contrastive texts

User study confirms improved insight and quality

Abstract

Contrastive explanations for understanding the behavior of black box models has gained a lot of attention recently as they provide potential for recourse. In this paper, we propose a method Contrastive Attributed explanations for Text (CAT) which provides contrastive explanations for natural language text data with a novel twist as we build and exploit attribute classifiers leading to more semantically meaningful explanations. To ensure that our contrastive generated text has the fewest possible edits with respect to the original text, while also being fluent and close to a human generated contrastive, we resort to a minimal perturbation approach regularized using a BERT language model and attribute classifiers trained on available attributes. We show through qualitative examples and a user study that our method not only conveys more insight because of these attributes, but also leads to better quality (contrastive) text. Quantitatively, we show that our method outperforms other state-of-the-art methods across four data sets on four benchmark metrics.