Multi-facet Universal Schema

Rohan Paul, Haw-Shiuan Chang, Andrew McCallum


Abstract
Universal schema (USchema) assumes that two sentence patterns that share the same entity pairs are similar to each other. This assumption is widely adopted for solving various types of relation extraction (RE) tasks. Nevertheless, each sentence pattern could contain multiple facets, and not every facet is similar to all the facets of another sentence pattern co-occurring with the same entity pair. To address the violation of the USchema assumption, we propose multi-facet universal schema that uses a neural model to represent each sentence pattern as multiple facet embeddings and encourage one of these facet embeddings to be close to that of another sentence pattern if they co-occur with the same entity pair. In our experiments, we demonstrate that multi-facet embeddings significantly outperform their single-facet embedding counterpart, compositional universal schema (CUSchema) (Verga et al., 2016), in distantly supervised relation extraction tasks. Moreover, we can also use multiple embeddings to detect the entailment relation between two sentence patterns when no manual label is available.
Anthology ID:
2021.eacl-main.77
Volume:
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume
Month:
April
Year:
2021
Address:
Online
Editors:
Paola Merlo, Jorg Tiedemann, Reut Tsarfaty
Venue:
EACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
909–919
Language:
URL:
https://aclanthology.org/2021.eacl-main.77
DOI:
10.18653/v1/2021.eacl-main.77
Bibkey:
Cite (ACL):
Rohan Paul, Haw-Shiuan Chang, and Andrew McCallum. 2021. Multi-facet Universal Schema. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 909–919, Online. Association for Computational Linguistics.
Cite (Informal):
Multi-facet Universal Schema (Paul et al., EACL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.eacl-main.77.pdf