Fetching the paper…

Neural Natural Language Inference Models Partially Embed Theories of Lexical Entailment and Negation · Around