@inproceedings{blevins-zettlemoyer-2020-moving,
title = "Moving Down the Long Tail of Word Sense Disambiguation with Gloss Informed Bi-encoders",
author = "Blevins, Terra and
Zettlemoyer, Luke",
editor = "Jurafsky, Dan and
Chai, Joyce and
Schluter, Natalie and
Tetreault, Joel",
booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
month = jul,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.acl-main.95",
doi = "10.18653/v1/2020.acl-main.95",
pages = "1006--1017",
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.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="blevins-zettlemoyer-2020-moving">
<titleInfo>
<title>Moving Down the Long Tail of Word Sense Disambiguation with Gloss Informed Bi-encoders</title>
</titleInfo>
<name type="personal">
<namePart type="given">Terra</namePart>
<namePart type="family">Blevins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Luke</namePart>
<namePart type="family">Zettlemoyer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dan</namePart>
<namePart type="family">Jurafsky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Chai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Natalie</namePart>
<namePart type="family">Schluter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joel</namePart>
<namePart type="family">Tetreault</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<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.</abstract>
<identifier type="citekey">blevins-zettlemoyer-2020-moving</identifier>
<identifier type="doi">10.18653/v1/2020.acl-main.95</identifier>
<location>
<url>https://aclanthology.org/2020.acl-main.95</url>
</location>
<part>
<date>2020-07</date>
<extent unit="page">
<start>1006</start>
<end>1017</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Moving Down the Long Tail of Word Sense Disambiguation with Gloss Informed Bi-encoders
%A Blevins, Terra
%A Zettlemoyer, Luke
%Y Jurafsky, Dan
%Y Chai, Joyce
%Y Schluter, Natalie
%Y Tetreault, Joel
%S Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
%D 2020
%8 July
%I Association for Computational Linguistics
%C Online
%F blevins-zettlemoyer-2020-moving
%X 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.
%R 10.18653/v1/2020.acl-main.95
%U https://aclanthology.org/2020.acl-main.95
%U https://doi.org/10.18653/v1/2020.acl-main.95
%P 1006-1017
Markdown (Informal)
[Moving Down the Long Tail of Word Sense Disambiguation with Gloss Informed Bi-encoders](https://aclanthology.org/2020.acl-main.95) (Blevins & Zettlemoyer, ACL 2020)
ACL