Towards Interpretable Natural Language Understanding with Explanations as Latent Variables

Abstract

Recently generating natural language explanations has shown very promising results in not only offering interpretable explanations but also providing additional information and supervision for prediction. However, existing approaches usually require a large set of human annotated explanations for training while collecting a large set of explanations is not only time consuming but also expensive. In this paper, we develop a general framework for interpretable natural language understanding that requires only a small set of human annotated explanations for training. Our framework treats natural language explanations as latent variables that model the underlying reasoning process of a neural model. We develop a variational EM framework for optimization where an explanation generation module and an explanation-augmented prediction module are alternatively optimized and mutually enhance each other. Moreover, we further propose an explanation-based self-training method under this framework for semi-supervised learning. It alternates between assigning pseudo-labels to unlabeled data and generating new explanations to iteratively improve each other. Experiments on two natural language understanding tasks demonstrate that our framework can not only make effective predictions in both supervised and semi-supervised settings, but also generate good natural language explanation.

NeurIPS 2020. The first three authors contribute equally

References in corpus (7)

• ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
• SemEval-2010 Task 8: Multi-Way Classification of Semantic Relations Between Pairs of Nominals
• Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints
• Language Models as Knowledge Bases?
• BERT Loses Patience: Fast and Robust Inference with Early Exit
• DeeBERT: Dynamic Early Exiting for Accelerating BERT Inference
• Learning from Explanations with Neural Execution Tree

Cited by in corpus (5)

• Towards Human-centered Explainable AI: A Survey of User Studies for Model Explanations
• LOREN: Logic-Regularized Reasoning for Interpretable Fact Verification
• You Can Do Better! If You Elaborate the Reason When Making Prediction
• SalKG: Learning From Knowledge Graph Explanations for Commonsense Reasoning
• Interpretable Multimodal Out-of-context Detection with Soft Logic Regularization