@inproceedings{khemchandani-etal-2021-exploiting,
title = "Exploiting Language Relatedness for Low Web-Resource Language Model Adaptation: {A}n {I}ndic Languages Study",
author = "Khemchandani, Yash and
Mehtani, Sarvesh and
Patil, Vaidehi and
Awasthi, Abhijeet and
Talukdar, Partha and
Sarawagi, Sunita",
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.105",
doi = "10.18653/v1/2021.acl-long.105",
pages = "1312--1323",
abstract = "Recent research in multilingual language models (LM) has demonstrated their ability to effectively handle multiple languages in a single model. This holds promise for low web-resource languages (LRL) as multilingual models can enable transfer of supervision from high resource languages to LRLs. However, incorporating a new language in an LM still remains a challenge, particularly for languages with limited corpora and in unseen scripts. In this paper we argue that relatedness among languages in a language family may be exploited to overcome some of the corpora limitations of LRLs, and propose RelateLM. We focus on Indian languages, and exploit relatedness along two dimensions: (1) script (since many Indic scripts originated from the Brahmic script), and (2) sentence structure. RelateLM uses transliteration to convert the unseen script of limited LRL text into the script of a Related Prominent Language (RPL) (Hindi in our case). While exploiting similar sentence structures, RelateLM utilizes readily available bilingual dictionaries to pseudo translate RPL text into LRL corpora. Experiments on multiple real-world benchmark datasets provide validation to our hypothesis that using a related language as pivot, along with transliteration and pseudo translation based data augmentation, can be an effective way to adapt LMs for LRLs, rather than direct training or pivoting through English.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="khemchandani-etal-2021-exploiting">
<titleInfo>
<title>Exploiting Language Relatedness for Low Web-Resource Language Model Adaptation: An Indic Languages Study</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yash</namePart>
<namePart type="family">Khemchandani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sarvesh</namePart>
<namePart type="family">Mehtani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vaidehi</namePart>
<namePart type="family">Patil</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abhijeet</namePart>
<namePart type="family">Awasthi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Partha</namePart>
<namePart type="family">Talukdar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunita</namePart>
<namePart type="family">Sarawagi</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>Recent research in multilingual language models (LM) has demonstrated their ability to effectively handle multiple languages in a single model. This holds promise for low web-resource languages (LRL) as multilingual models can enable transfer of supervision from high resource languages to LRLs. However, incorporating a new language in an LM still remains a challenge, particularly for languages with limited corpora and in unseen scripts. In this paper we argue that relatedness among languages in a language family may be exploited to overcome some of the corpora limitations of LRLs, and propose RelateLM. We focus on Indian languages, and exploit relatedness along two dimensions: (1) script (since many Indic scripts originated from the Brahmic script), and (2) sentence structure. RelateLM uses transliteration to convert the unseen script of limited LRL text into the script of a Related Prominent Language (RPL) (Hindi in our case). While exploiting similar sentence structures, RelateLM utilizes readily available bilingual dictionaries to pseudo translate RPL text into LRL corpora. Experiments on multiple real-world benchmark datasets provide validation to our hypothesis that using a related language as pivot, along with transliteration and pseudo translation based data augmentation, can be an effective way to adapt LMs for LRLs, rather than direct training or pivoting through English.</abstract>
<identifier type="citekey">khemchandani-etal-2021-exploiting</identifier>
<identifier type="doi">10.18653/v1/2021.acl-long.105</identifier>
<location>
<url>https://aclanthology.org/2021.acl-long.105</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>1312</start>
<end>1323</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Exploiting Language Relatedness for Low Web-Resource Language Model Adaptation: An Indic Languages Study
%A Khemchandani, Yash
%A Mehtani, Sarvesh
%A Patil, Vaidehi
%A Awasthi, Abhijeet
%A Talukdar, Partha
%A Sarawagi, Sunita
%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 khemchandani-etal-2021-exploiting
%X Recent research in multilingual language models (LM) has demonstrated their ability to effectively handle multiple languages in a single model. This holds promise for low web-resource languages (LRL) as multilingual models can enable transfer of supervision from high resource languages to LRLs. However, incorporating a new language in an LM still remains a challenge, particularly for languages with limited corpora and in unseen scripts. In this paper we argue that relatedness among languages in a language family may be exploited to overcome some of the corpora limitations of LRLs, and propose RelateLM. We focus on Indian languages, and exploit relatedness along two dimensions: (1) script (since many Indic scripts originated from the Brahmic script), and (2) sentence structure. RelateLM uses transliteration to convert the unseen script of limited LRL text into the script of a Related Prominent Language (RPL) (Hindi in our case). While exploiting similar sentence structures, RelateLM utilizes readily available bilingual dictionaries to pseudo translate RPL text into LRL corpora. Experiments on multiple real-world benchmark datasets provide validation to our hypothesis that using a related language as pivot, along with transliteration and pseudo translation based data augmentation, can be an effective way to adapt LMs for LRLs, rather than direct training or pivoting through English.
%R 10.18653/v1/2021.acl-long.105
%U https://aclanthology.org/2021.acl-long.105
%U https://doi.org/10.18653/v1/2021.acl-long.105
%P 1312-1323
Markdown (Informal)
[Exploiting Language Relatedness for Low Web-Resource Language Model Adaptation: An Indic Languages Study](https://aclanthology.org/2021.acl-long.105) (Khemchandani et al., ACL-IJCNLP 2021)
ACL