Projects per year
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 non-directional 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.
|Title of host publication||Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing|
|Editors||Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih|
|Place of Publication||Stroudsburg, PA, United States|
|Number of pages||11|
|Publication status||Published - 7 Nov 2021|
|Event||2021 Conference on Empirical Methods in Natural Language Processing - Punta Cana, Dominican Republic|
Duration: 7 Nov 2021 → 11 Nov 2021
|Conference||2021 Conference on Empirical Methods in Natural Language Processing|
|Abbreviated title||EMNLP 2021|
|Period||7/11/21 → 11/11/21|
FingerprintDive into the research topics of 'Multivalent Entailment Graphs for Question Answering'. Together they form a unique fingerprint.
- 1 Active
1/08/17 → 31/01/23