Research Area:  Machine Learning
Deep learning models perform poorly on tasks that require commonsense reasoning, which often necessitates some form of world-knowledge or reasoning over information not immediately present in the input. We collect human explanations for commonsense reasoning in the form of natural language sequences and highlighted annotations in a new dataset called Common Sense Explanations (CoS-E). We use CoS-E to train language models to automatically generate explanations that can be used during training and inference in a novel Commonsense Auto-Generated Explanation (CAGE) framework. CAGE improves the state-of-the-art by 10% on the challenging Commonsense QA task. We further study commonsense reasoning in DNNs using both human and auto-generated explanations including transfer to out-of-domain tasks. Empirical results indicate that we can effectively leverage language models for commonsense reasoning.
Keywords:  
Common Sense Explanations
Commonsense Auto-Generated Explanation
Computation
Language
Commonsense Reasoning
Author(s) Name:  Nazneen Fatema Rajani, Bryan McCann, Caiming Xiong, Richard Socher
Journal name:   Computation and Language
Conferrence name:  
Publisher name:  arXiv.1906.02361
DOI:  10.48550/arXiv.1906.02361
Volume Information:  
Paper Link:   https://arxiv.org/abs/1906.02361