@inproceedings{vulic-etal-2021-lexfit,
title = "{L}ex{F}it: Lexical Fine-Tuning of Pretrained Language Models",
author = "Vuli{\'c}, Ivan and
Ponti, Edoardo Maria and
Korhonen, Anna and
Glava{\v{s}}, Goran",
editor = "Zong, Chengqing and
Xia, Fei and
Li, Wenjie and
Navigli, Roberto",
booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.acl-long.410",
doi = "10.18653/v1/2021.acl-long.410",
pages = "5269--5283",
abstract = "Transformer-based language models (LMs) pretrained on large text collections implicitly store a wealth of lexical semantic knowledge, but it is non-trivial to extract that knowledge effectively from their parameters. Inspired by prior work on semantic specialization of static word embedding (WE) models, we show that it is possible to expose and enrich lexical knowledge from the LMs, that is, to specialize them to serve as effective and universal {``}decontextualized{''} word encoders even when fed input words {``}in isolation{''} (i.e., without any context). Their transformation into such word encoders is achieved through a simple and efficient lexical fine-tuning procedure (termed LexFit) based on dual-encoder network structures. Further, we show that LexFit can yield effective word encoders even with limited lexical supervision and, via cross-lingual transfer, in different languages without any readily available external knowledge. Our evaluation over four established, structurally different lexical-level tasks in 8 languages indicates the superiority of LexFit-based WEs over standard static WEs (e.g., fastText) and WEs from vanilla LMs. Other extensive experiments and ablation studies further profile the LexFit framework, and indicate best practices and performance variations across LexFit variants, languages, and lexical tasks, also directly questioning the usefulness of traditional WE models in the era of large neural models.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="vulic-etal-2021-lexfit">
<titleInfo>
<title>LexFit: Lexical Fine-Tuning of Pretrained Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="family">Vulić</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Edoardo</namePart>
<namePart type="given">Maria</namePart>
<namePart type="family">Ponti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Korhonen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Goran</namePart>
<namePart type="family">Glavaš</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chengqing</namePart>
<namePart type="family">Zong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fei</namePart>
<namePart type="family">Xia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenjie</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roberto</namePart>
<namePart type="family">Navigli</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>Transformer-based language models (LMs) pretrained on large text collections implicitly store a wealth of lexical semantic knowledge, but it is non-trivial to extract that knowledge effectively from their parameters. Inspired by prior work on semantic specialization of static word embedding (WE) models, we show that it is possible to expose and enrich lexical knowledge from the LMs, that is, to specialize them to serve as effective and universal “decontextualized” word encoders even when fed input words “in isolation” (i.e., without any context). Their transformation into such word encoders is achieved through a simple and efficient lexical fine-tuning procedure (termed LexFit) based on dual-encoder network structures. Further, we show that LexFit can yield effective word encoders even with limited lexical supervision and, via cross-lingual transfer, in different languages without any readily available external knowledge. Our evaluation over four established, structurally different lexical-level tasks in 8 languages indicates the superiority of LexFit-based WEs over standard static WEs (e.g., fastText) and WEs from vanilla LMs. Other extensive experiments and ablation studies further profile the LexFit framework, and indicate best practices and performance variations across LexFit variants, languages, and lexical tasks, also directly questioning the usefulness of traditional WE models in the era of large neural models.</abstract>
<identifier type="citekey">vulic-etal-2021-lexfit</identifier>
<identifier type="doi">10.18653/v1/2021.acl-long.410</identifier>
<location>
<url>https://aclanthology.org/2021.acl-long.410</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>5269</start>
<end>5283</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LexFit: Lexical Fine-Tuning of Pretrained Language Models
%A Vulić, Ivan
%A Ponti, Edoardo Maria
%A Korhonen, Anna
%A Glavaš, Goran
%Y Zong, Chengqing
%Y Xia, Fei
%Y Li, Wenjie
%Y Navigli, Roberto
%S Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
%D 2021
%8 August
%I Association for Computational Linguistics
%C Online
%F vulic-etal-2021-lexfit
%X Transformer-based language models (LMs) pretrained on large text collections implicitly store a wealth of lexical semantic knowledge, but it is non-trivial to extract that knowledge effectively from their parameters. Inspired by prior work on semantic specialization of static word embedding (WE) models, we show that it is possible to expose and enrich lexical knowledge from the LMs, that is, to specialize them to serve as effective and universal “decontextualized” word encoders even when fed input words “in isolation” (i.e., without any context). Their transformation into such word encoders is achieved through a simple and efficient lexical fine-tuning procedure (termed LexFit) based on dual-encoder network structures. Further, we show that LexFit can yield effective word encoders even with limited lexical supervision and, via cross-lingual transfer, in different languages without any readily available external knowledge. Our evaluation over four established, structurally different lexical-level tasks in 8 languages indicates the superiority of LexFit-based WEs over standard static WEs (e.g., fastText) and WEs from vanilla LMs. Other extensive experiments and ablation studies further profile the LexFit framework, and indicate best practices and performance variations across LexFit variants, languages, and lexical tasks, also directly questioning the usefulness of traditional WE models in the era of large neural models.
%R 10.18653/v1/2021.acl-long.410
%U https://aclanthology.org/2021.acl-long.410
%U https://doi.org/10.18653/v1/2021.acl-long.410
%P 5269-5283
Markdown (Informal)
[LexFit: Lexical Fine-Tuning of Pretrained Language Models](https://aclanthology.org/2021.acl-long.410) (Vulić et al., ACL-IJCNLP 2021)
ACL
- Ivan Vulić, Edoardo Maria Ponti, Anna Korhonen, and Goran Glavaš. 2021. LexFit: Lexical Fine-Tuning of Pretrained Language Models. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 5269–5283, Online. Association for Computational Linguistics.