Multivalent Entailment Graphs for Question Answering
Abstract
Drawing inferences between open-domain natural language predicates is a necessity for true language understanding. There has been much progress in unsupervised learning of entailment graphs for this purpose. We make three contributions: (1) we reinterpret the Distributional Inclusion Hypothesis to model entailment between predicates of different valencies, like DEFEAT(Biden, Trump) entails WIN(Biden); (2) we actualize this theory by learning unsupervised Multivalent Entailment Graphs of open-domain predicates; and (3) we demonstrate the capabilities of these graphs on a novel question answering task. We show that directional entailment is more helpful for inference than bidirectional similarity on questions of fine-grained semantics. We also show that drawing on evidence across valencies answers more questions than by using only the same valency evidence.
Cite
@article{arxiv.2104.07846,
title = {Multivalent Entailment Graphs for Question Answering},
author = {Nick McKenna and Liane Guillou and Mohammad Javad Hosseini and Sander Bijl de Vroe and Mark Johnson and Mark Steedman},
journal= {arXiv preprint arXiv:2104.07846},
year = {2021}
}
Comments
Accepted to EMNLP 2021