We propose a process for investigating the extent to which sentence representations arising from neural machine translation (NMT) systems encode distinct semantic phenomena. We use these representations as features to train a natural language inference (NLI) classifier based on datasets recast from existing semantic annotations. In applying this process to a representative NMT system, we find its encoder appears most suited to supporting inferences at the syntax-semantics interface, as compared to anaphora resolution requiring world-knowledge. We conclude with a discussion on the merits and potential deficiencies of the existing process, and how it may be improved and extended as a broader framework for evaluating semantic coverage.
@article{arxiv.1804.09779,
title = {On the Evaluation of Semantic Phenomena in Neural Machine Translation Using Natural Language Inference},
author = {Adam Poliak and Yonatan Belinkov and James Glass and Benjamin Van Durme},
journal= {arXiv preprint arXiv:1804.09779},
year = {2018}
}