KACE: Generating Knowledge Aware Contrastive Explanations for Natural Language Inference

In order to better understand the reason behind model behaviors (i.e., making predictions), most recent work has exploited generative models to provide complementary explanations. However, existing approaches in natural language processing (NLP) mainly focus on "WHY A" rather than contrastive "WHY A NOT B", which is shown to be able to better distinguish confusing candidates and improve model performance in other research fields. In this paper, we focus on generating Contrastive Explanations with counterfactual examples in NLI and propose a novel Knowledge-Aware generation framework (KACE). Specifically, we first identify rationales (i.e., key phrases) from input sentences, and use them as key perturbations for generating counterfactual examples. After obtaining qualified counterfactual examples, we take them along with original examples and external knowledge as input, and employ a knowledge-aware generative pre-trained language model to generate contrastive explanations. Experimental results show that contrastive explanations are beneficial to clarify the difference between predicted answer and other answer options. Moreover, we train an BERT-large based NLI model enhanced with contrastive explanations and achieve an accuracy of 91.9% on SNLI, gaining an improvement of 5.7% against ETPA ("Explain-Then-Predict-Attention") and 0.6% against NILE ("WHY A").