@inproceedings{jeong-etal-2024-bootstrap,
title = "Bootstrap Your Own {PLM}: Boosting Semantic Features of {PLM}s for Unsuperivsed Contrastive Learning",
author = "Jeong, Yoo Hyun and
Han, Myeong Soo and
Chae, Dong-Kyu",
editor = "Graham, Yvette and
Purver, Matthew",
booktitle = "Findings of the Association for Computational Linguistics: EACL 2024",
month = mar,
year = "2024",
address = "St. Julian{'}s, Malta",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-eacl.38",
pages = "560--569",
abstract = "This paper aims to investigate the possibility of exploiting original semantic features of PLMs (pre-trained language models) during contrastive learning in the context of SRL (sentence representation learning). In the context of feature modification, we identified a method called IFM (implicit feature modification), which reduces the tendency of contrastive models for VRL (visual representation learning) to rely on feature-suppressing shortcut solutions. We observed that IFM did not work well for SRL, which may be due to differences between the nature of VRL and SRL. We propose BYOP, which boosts well-represented features, taking the opposite idea of IFM, under the assumption that SimCSE{'}s dropout-noise-based augmentation may be too simple to modify high-level semantic features, and that the features learned by PLMs are semantically meaningful and should be boosted, rather than removed. Extensive experiments lend credence to the logic of BYOP, which considers the nature of SRL.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jeong-etal-2024-bootstrap">
<titleInfo>
<title>Bootstrap Your Own PLM: Boosting Semantic Features of PLMs for Unsuperivsed Contrastive Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoo</namePart>
<namePart type="given">Hyun</namePart>
<namePart type="family">Jeong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Myeong</namePart>
<namePart type="given">Soo</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dong-Kyu</namePart>
<namePart type="family">Chae</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yvette</namePart>
<namePart type="family">Graham</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matthew</namePart>
<namePart type="family">Purver</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">St. Julian’s, Malta</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper aims to investigate the possibility of exploiting original semantic features of PLMs (pre-trained language models) during contrastive learning in the context of SRL (sentence representation learning). In the context of feature modification, we identified a method called IFM (implicit feature modification), which reduces the tendency of contrastive models for VRL (visual representation learning) to rely on feature-suppressing shortcut solutions. We observed that IFM did not work well for SRL, which may be due to differences between the nature of VRL and SRL. We propose BYOP, which boosts well-represented features, taking the opposite idea of IFM, under the assumption that SimCSE’s dropout-noise-based augmentation may be too simple to modify high-level semantic features, and that the features learned by PLMs are semantically meaningful and should be boosted, rather than removed. Extensive experiments lend credence to the logic of BYOP, which considers the nature of SRL.</abstract>
<identifier type="citekey">jeong-etal-2024-bootstrap</identifier>
<location>
<url>https://aclanthology.org/2024.findings-eacl.38</url>
</location>
<part>
<date>2024-03</date>
<extent unit="page">
<start>560</start>
<end>569</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Bootstrap Your Own PLM: Boosting Semantic Features of PLMs for Unsuperivsed Contrastive Learning
%A Jeong, Yoo Hyun
%A Han, Myeong Soo
%A Chae, Dong-Kyu
%Y Graham, Yvette
%Y Purver, Matthew
%S Findings of the Association for Computational Linguistics: EACL 2024
%D 2024
%8 March
%I Association for Computational Linguistics
%C St. Julian’s, Malta
%F jeong-etal-2024-bootstrap
%X This paper aims to investigate the possibility of exploiting original semantic features of PLMs (pre-trained language models) during contrastive learning in the context of SRL (sentence representation learning). In the context of feature modification, we identified a method called IFM (implicit feature modification), which reduces the tendency of contrastive models for VRL (visual representation learning) to rely on feature-suppressing shortcut solutions. We observed that IFM did not work well for SRL, which may be due to differences between the nature of VRL and SRL. We propose BYOP, which boosts well-represented features, taking the opposite idea of IFM, under the assumption that SimCSE’s dropout-noise-based augmentation may be too simple to modify high-level semantic features, and that the features learned by PLMs are semantically meaningful and should be boosted, rather than removed. Extensive experiments lend credence to the logic of BYOP, which considers the nature of SRL.
%U https://aclanthology.org/2024.findings-eacl.38
%P 560-569
Markdown (Informal)
[Bootstrap Your Own PLM: Boosting Semantic Features of PLMs for Unsuperivsed Contrastive Learning](https://aclanthology.org/2024.findings-eacl.38) (Jeong et al., Findings 2024)
ACL