Towards Robustifying NLI Models Against Lexical Dataset Biases

Abstract

Publication
In the 2020 Annual Conference of the Association for Computational Linguistics
Date