Moving Down the Long Tail of Word Sense Disambiguation with Gloss Informed Bi-encoders

Terra Blevins, Luke Zettlemoyer


Abstract
A major obstacle in Word Sense Disambiguation (WSD) is that word senses are not uniformly distributed, causing existing models to generally perform poorly on senses that are either rare or unseen during training. We propose a bi-encoder model that independently embeds (1) the target word with its surrounding context and (2) the dictionary definition, or gloss, of each sense. The encoders are jointly optimized in the same representation space, so that sense disambiguation can be performed by finding the nearest sense embedding for each target word embedding. Our system outperforms previous state-of-the-art models on English all-words WSD; these gains predominantly come from improved performance on rare senses, leading to a 31.1% error reduction on less frequent senses over prior work. This demonstrates that rare senses can be more effectively disambiguated by modeling their definitions.
Anthology ID:
2020.acl-main.95
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1006–1017
Language:
URL:
https://aclanthology.org/2020.acl-main.95
DOI:
10.18653/v1/2020.acl-main.95
Bibkey:
Cite (ACL):
Terra Blevins and Luke Zettlemoyer. 2020. Moving Down the Long Tail of Word Sense Disambiguation with Gloss Informed Bi-encoders. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1006–1017, Online. Association for Computational Linguistics.
Cite (Informal):
Moving Down the Long Tail of Word Sense Disambiguation with Gloss Informed Bi-encoders (Blevins & Zettlemoyer, ACL 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.acl-main.95.pdf
Video:
 http://slideslive.com/38928744
Data
Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison