@inproceedings{das-etal-2023-improving,
title = "Improving Pretraining Techniques for Code-Switched {NLP}",
author = "Das, Richeek and
Ranjan, Sahasra and
Pathak, Shreya and
Jyothi, Preethi",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.66",
doi = "10.18653/v1/2023.acl-long.66",
pages = "1176--1191",
abstract = "Pretrained models are a mainstay in modern NLP applications. Pretraining requires access to large volumes of unlabeled text. While monolingual text is readily available for many of the world{'}s languages, access to large quantities of code-switched text (i.e., text with tokens of multiple languages interspersed within a sentence) is much more scarce. Given this resource constraint, the question of how pretraining using limited amounts of code-switched text could be altered to improve performance for code-switched NLP becomes important to tackle. In this paper, we explore different masked language modeling (MLM) pretraining techniques for code-switched text that are cognizant of language boundaries prior to masking. The language identity of the tokens can either come from human annotators, trained language classifiers, or simple relative frequency-based estimates. We also present an MLM variant by introducing a residual connection from an earlier layer in the pretrained model that uniformly boosts performance on downstream tasks. Experiments on two downstream tasks, Question Answering (QA) and Sentiment Analysis (SA), involving four code-switched language pairs (Hindi-English, Spanish-English, Tamil-English, Malayalam-English) yield relative improvements of up to 5.8 and 2.7 F1 scores on QA (Hindi-English) and SA (Tamil-English), respectively, compared to standard pretraining techniques. To understand our task improvements better, we use a series of probes to study what additional information is encoded by our pretraining techniques and also introduce an auxiliary loss function that explicitly models language identification to further aid the residual MLM variants.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="das-etal-2023-improving">
<titleInfo>
<title>Improving Pretraining Techniques for Code-Switched NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Richeek</namePart>
<namePart type="family">Das</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sahasra</namePart>
<namePart type="family">Ranjan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shreya</namePart>
<namePart type="family">Pathak</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Preethi</namePart>
<namePart type="family">Jyothi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Pretrained models are a mainstay in modern NLP applications. Pretraining requires access to large volumes of unlabeled text. While monolingual text is readily available for many of the world’s languages, access to large quantities of code-switched text (i.e., text with tokens of multiple languages interspersed within a sentence) is much more scarce. Given this resource constraint, the question of how pretraining using limited amounts of code-switched text could be altered to improve performance for code-switched NLP becomes important to tackle. In this paper, we explore different masked language modeling (MLM) pretraining techniques for code-switched text that are cognizant of language boundaries prior to masking. The language identity of the tokens can either come from human annotators, trained language classifiers, or simple relative frequency-based estimates. We also present an MLM variant by introducing a residual connection from an earlier layer in the pretrained model that uniformly boosts performance on downstream tasks. Experiments on two downstream tasks, Question Answering (QA) and Sentiment Analysis (SA), involving four code-switched language pairs (Hindi-English, Spanish-English, Tamil-English, Malayalam-English) yield relative improvements of up to 5.8 and 2.7 F1 scores on QA (Hindi-English) and SA (Tamil-English), respectively, compared to standard pretraining techniques. To understand our task improvements better, we use a series of probes to study what additional information is encoded by our pretraining techniques and also introduce an auxiliary loss function that explicitly models language identification to further aid the residual MLM variants.</abstract>
<identifier type="citekey">das-etal-2023-improving</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.66</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.66</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>1176</start>
<end>1191</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Improving Pretraining Techniques for Code-Switched NLP
%A Das, Richeek
%A Ranjan, Sahasra
%A Pathak, Shreya
%A Jyothi, Preethi
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F das-etal-2023-improving
%X Pretrained models are a mainstay in modern NLP applications. Pretraining requires access to large volumes of unlabeled text. While monolingual text is readily available for many of the world’s languages, access to large quantities of code-switched text (i.e., text with tokens of multiple languages interspersed within a sentence) is much more scarce. Given this resource constraint, the question of how pretraining using limited amounts of code-switched text could be altered to improve performance for code-switched NLP becomes important to tackle. In this paper, we explore different masked language modeling (MLM) pretraining techniques for code-switched text that are cognizant of language boundaries prior to masking. The language identity of the tokens can either come from human annotators, trained language classifiers, or simple relative frequency-based estimates. We also present an MLM variant by introducing a residual connection from an earlier layer in the pretrained model that uniformly boosts performance on downstream tasks. Experiments on two downstream tasks, Question Answering (QA) and Sentiment Analysis (SA), involving four code-switched language pairs (Hindi-English, Spanish-English, Tamil-English, Malayalam-English) yield relative improvements of up to 5.8 and 2.7 F1 scores on QA (Hindi-English) and SA (Tamil-English), respectively, compared to standard pretraining techniques. To understand our task improvements better, we use a series of probes to study what additional information is encoded by our pretraining techniques and also introduce an auxiliary loss function that explicitly models language identification to further aid the residual MLM variants.
%R 10.18653/v1/2023.acl-long.66
%U https://aclanthology.org/2023.acl-long.66
%U https://doi.org/10.18653/v1/2023.acl-long.66
%P 1176-1191
Markdown (Informal)
[Improving Pretraining Techniques for Code-Switched NLP](https://aclanthology.org/2023.acl-long.66) (Das et al., ACL 2023)
ACL
- Richeek Das, Sahasra Ranjan, Shreya Pathak, and Preethi Jyothi. 2023. Improving Pretraining Techniques for Code-Switched NLP. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1176–1191, Toronto, Canada. Association for Computational Linguistics.