@inproceedings{yava-etal-2024-improving,
title = "Improving Word Sense Induction through Adversarial Forgetting of Morphosyntactic Information",
author = "Yavas, Deniz Ekin and
Bernard, Timoth{\'e}e and
Kallmeyer, Laura and
Crabb{\'e}, Beno{\^i}t",
editor = "Bollegala, Danushka and
Shwartz, Vered",
booktitle = "Proceedings of the 13th Joint Conference on Lexical and Computational Semantics (*SEM 2024)",
month = jun,
year = "2024",
address = "Mexico City, Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.starsem-1.19/",
doi = "10.18653/v1/2024.starsem-1.19",
pages = "238--251",
abstract = "This paper addresses the problem of word sense induction (WSI) via clustering of word embeddings. It starts from the hypothesis that contextualized word representations obtained from pre-trained language models (LMs), while being a valuable source for WSI, encode more information than what is necessary for the identification of word senses and some of this information affect the performance negatively in unsupervised settings. We investigate whether using contextualized representations that are invariant to these {\textquoteleft}nuisance features' can increase WSI performance. For this purpose, we propose an adaptation of the adversarial training framework proposed by Jaiswal et al. (2020) to erase specific information from the representations of LMs, thereby creating feature-invariant representations. We experiment with erasing (i) morphological and (ii) syntactic features. The results of subsequent clustering for WSI show that these features indeed act like noise: Using feature-invariant representations, compared to using the original representations, increases clustering-based WSI performance. Furthermore, we provide an in-depth analysis of how the information about the syntactic and morphological features of words relate to and affect WSI performance."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yava-etal-2024-improving">
<titleInfo>
<title>Improving Word Sense Induction through Adversarial Forgetting of Morphosyntactic Information</title>
</titleInfo>
<name type="personal">
<namePart type="given">Deniz</namePart>
<namePart type="given">Ekin</namePart>
<namePart type="family">Yavas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Timothée</namePart>
<namePart type="family">Bernard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Laura</namePart>
<namePart type="family">Kallmeyer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Benoît</namePart>
<namePart type="family">Crabbé</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 13th Joint Conference on Lexical and Computational Semantics (*SEM 2024)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Danushka</namePart>
<namePart type="family">Bollegala</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vered</namePart>
<namePart type="family">Shwartz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Mexico City, Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper addresses the problem of word sense induction (WSI) via clustering of word embeddings. It starts from the hypothesis that contextualized word representations obtained from pre-trained language models (LMs), while being a valuable source for WSI, encode more information than what is necessary for the identification of word senses and some of this information affect the performance negatively in unsupervised settings. We investigate whether using contextualized representations that are invariant to these ‘nuisance features’ can increase WSI performance. For this purpose, we propose an adaptation of the adversarial training framework proposed by Jaiswal et al. (2020) to erase specific information from the representations of LMs, thereby creating feature-invariant representations. We experiment with erasing (i) morphological and (ii) syntactic features. The results of subsequent clustering for WSI show that these features indeed act like noise: Using feature-invariant representations, compared to using the original representations, increases clustering-based WSI performance. Furthermore, we provide an in-depth analysis of how the information about the syntactic and morphological features of words relate to and affect WSI performance.</abstract>
<identifier type="citekey">yava-etal-2024-improving</identifier>
<identifier type="doi">10.18653/v1/2024.starsem-1.19</identifier>
<location>
<url>https://aclanthology.org/2024.starsem-1.19/</url>
</location>
<part>
<date>2024-06</date>
<extent unit="page">
<start>238</start>
<end>251</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Improving Word Sense Induction through Adversarial Forgetting of Morphosyntactic Information
%A Yavas, Deniz Ekin
%A Bernard, Timothée
%A Kallmeyer, Laura
%A Crabbé, Benoît
%Y Bollegala, Danushka
%Y Shwartz, Vered
%S Proceedings of the 13th Joint Conference on Lexical and Computational Semantics (*SEM 2024)
%D 2024
%8 June
%I Association for Computational Linguistics
%C Mexico City, Mexico
%F yava-etal-2024-improving
%X This paper addresses the problem of word sense induction (WSI) via clustering of word embeddings. It starts from the hypothesis that contextualized word representations obtained from pre-trained language models (LMs), while being a valuable source for WSI, encode more information than what is necessary for the identification of word senses and some of this information affect the performance negatively in unsupervised settings. We investigate whether using contextualized representations that are invariant to these ‘nuisance features’ can increase WSI performance. For this purpose, we propose an adaptation of the adversarial training framework proposed by Jaiswal et al. (2020) to erase specific information from the representations of LMs, thereby creating feature-invariant representations. We experiment with erasing (i) morphological and (ii) syntactic features. The results of subsequent clustering for WSI show that these features indeed act like noise: Using feature-invariant representations, compared to using the original representations, increases clustering-based WSI performance. Furthermore, we provide an in-depth analysis of how the information about the syntactic and morphological features of words relate to and affect WSI performance.
%R 10.18653/v1/2024.starsem-1.19
%U https://aclanthology.org/2024.starsem-1.19/
%U https://doi.org/10.18653/v1/2024.starsem-1.19
%P 238-251
Markdown (Informal)
[Improving Word Sense Induction through Adversarial Forgetting of Morphosyntactic Information](https://aclanthology.org/2024.starsem-1.19/) (Yavas et al., *SEM 2024)
ACL