@inproceedings{karidi-etal-2021-putting,
title = "Putting Words in {BERT}`s Mouth: Navigating Contextualized Vector Spaces with Pseudowords",
author = "Karidi, Taelin and
Zhou, Yichu and
Schneider, Nathan and
Abend, Omri and
Srikumar, Vivek",
editor = "Moens, Marie-Francine and
Huang, Xuanjing and
Specia, Lucia and
Yih, Scott Wen-tau",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-main.806/",
doi = "10.18653/v1/2021.emnlp-main.806",
pages = "10300--10313",
abstract = "We present a method for exploring regions around individual points in a contextualized vector space (particularly, BERT space), as a way to investigate how these regions correspond to word senses. By inducing a contextualized {\textquotedblleft}pseudoword{\textquotedblright} vector as a stand-in for a static embedding in the input layer, and then performing masked prediction of a word in the sentence, we are able to investigate the geometry of the BERT-space in a controlled manner around individual instances. Using our method on a set of carefully constructed sentences targeting highly ambiguous English words, we find substantial regularity in the contextualized space, with regions that correspond to distinct word senses; but between these regions there are occasionally {\textquotedblleft}sense voids{\textquotedblright}{---}regions that do not correspond to any intelligible sense."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="karidi-etal-2021-putting">
<titleInfo>
<title>Putting Words in BERT‘s Mouth: Navigating Contextualized Vector Spaces with Pseudowords</title>
</titleInfo>
<name type="personal">
<namePart type="given">Taelin</namePart>
<namePart type="family">Karidi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yichu</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nathan</namePart>
<namePart type="family">Schneider</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Omri</namePart>
<namePart type="family">Abend</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marie-Francine</namePart>
<namePart type="family">Moens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanjing</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Specia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Scott</namePart>
<namePart type="given">Wen-tau</namePart>
<namePart type="family">Yih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online and Punta Cana, Dominican Republic</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We present a method for exploring regions around individual points in a contextualized vector space (particularly, BERT space), as a way to investigate how these regions correspond to word senses. By inducing a contextualized “pseudoword” vector as a stand-in for a static embedding in the input layer, and then performing masked prediction of a word in the sentence, we are able to investigate the geometry of the BERT-space in a controlled manner around individual instances. Using our method on a set of carefully constructed sentences targeting highly ambiguous English words, we find substantial regularity in the contextualized space, with regions that correspond to distinct word senses; but between these regions there are occasionally “sense voids”—regions that do not correspond to any intelligible sense.</abstract>
<identifier type="citekey">karidi-etal-2021-putting</identifier>
<identifier type="doi">10.18653/v1/2021.emnlp-main.806</identifier>
<location>
<url>https://aclanthology.org/2021.emnlp-main.806/</url>
</location>
<part>
<date>2021-11</date>
<extent unit="page">
<start>10300</start>
<end>10313</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Putting Words in BERT‘s Mouth: Navigating Contextualized Vector Spaces with Pseudowords
%A Karidi, Taelin
%A Zhou, Yichu
%A Schneider, Nathan
%A Abend, Omri
%A Srikumar, Vivek
%Y Moens, Marie-Francine
%Y Huang, Xuanjing
%Y Specia, Lucia
%Y Yih, Scott Wen-tau
%S Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
%D 2021
%8 November
%I Association for Computational Linguistics
%C Online and Punta Cana, Dominican Republic
%F karidi-etal-2021-putting
%X We present a method for exploring regions around individual points in a contextualized vector space (particularly, BERT space), as a way to investigate how these regions correspond to word senses. By inducing a contextualized “pseudoword” vector as a stand-in for a static embedding in the input layer, and then performing masked prediction of a word in the sentence, we are able to investigate the geometry of the BERT-space in a controlled manner around individual instances. Using our method on a set of carefully constructed sentences targeting highly ambiguous English words, we find substantial regularity in the contextualized space, with regions that correspond to distinct word senses; but between these regions there are occasionally “sense voids”—regions that do not correspond to any intelligible sense.
%R 10.18653/v1/2021.emnlp-main.806
%U https://aclanthology.org/2021.emnlp-main.806/
%U https://doi.org/10.18653/v1/2021.emnlp-main.806
%P 10300-10313
Markdown (Informal)
[Putting Words in BERT’s Mouth: Navigating Contextualized Vector Spaces with Pseudowords](https://aclanthology.org/2021.emnlp-main.806/) (Karidi et al., EMNLP 2021)
ACL