@inproceedings{kumar-etal-2026-sentence,
title = "Sentence-Level Back-Transliteration of {R}omanized {I}ndian Languages: Performance Analysis and Challenges",
author = "Kumar, Saurabh and
Kakadiya, Dhruvkumar Babubhai and
Singh, Sanasam Ranbir and
Nandi, Sukumar",
editor = "Piperidis, Stelios and
Bel, N{\'u}ria and
van den Heuvel, Henk and
Ide, Nancy and
Krek, Simon and
Toral, Antonio",
booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "ELRA Language Resource Association",
url = "https://aclanthology.org/2026.lrec-1.61/",
doi = "10.63317/2fe72eyjbmj3",
pages = "818--827",
abstract = "The widespread use of Romanized text for Indian languages, particularly on social media platforms, poses significant challenges for natural language processing due to the lack of standardized orthography and the presence of contextual ambiguities. In this study, we explore sentence-level back-transliteration for 13 Indian languages, focusing on addressing the limitations of word-level models that fail to capture contextual dependencies. We evaluate state-of-the-art models, including fine-tuned LLaMA, mT5, and Multilingual Transformer models, comparing their performance against the baseline IndicXlit model. In addition, we conduct a comprehensive error analysis to gain deeper insights into model performance. Our results demonstrate that fine-tuned LLaMA and the proposed IndiXform model, specifically designed to leverage sentence-level context, significantly outperform zero-shot LLaMA and the IndicXlit baseline. These findings provide valuable insights into handling contextual ambiguities and enhancing the accuracy of back-transliteration systems for Indian languages."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kumar-etal-2026-sentence">
<titleInfo>
<title>Sentence-Level Back-Transliteration of Romanized Indian Languages: Performance Analysis and Challenges</title>
</titleInfo>
<name type="personal">
<namePart type="given">Saurabh</namePart>
<namePart type="family">Kumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dhruvkumar</namePart>
<namePart type="given">Babubhai</namePart>
<namePart type="family">Kakadiya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sanasam</namePart>
<namePart type="given">Ranbir</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sukumar</namePart>
<namePart type="family">Nandi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Núria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henk</namePart>
<namePart type="family">van den Heuvel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nancy</namePart>
<namePart type="family">Ide</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Krek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Toral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resource Association</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The widespread use of Romanized text for Indian languages, particularly on social media platforms, poses significant challenges for natural language processing due to the lack of standardized orthography and the presence of contextual ambiguities. In this study, we explore sentence-level back-transliteration for 13 Indian languages, focusing on addressing the limitations of word-level models that fail to capture contextual dependencies. We evaluate state-of-the-art models, including fine-tuned LLaMA, mT5, and Multilingual Transformer models, comparing their performance against the baseline IndicXlit model. In addition, we conduct a comprehensive error analysis to gain deeper insights into model performance. Our results demonstrate that fine-tuned LLaMA and the proposed IndiXform model, specifically designed to leverage sentence-level context, significantly outperform zero-shot LLaMA and the IndicXlit baseline. These findings provide valuable insights into handling contextual ambiguities and enhancing the accuracy of back-transliteration systems for Indian languages.</abstract>
<identifier type="citekey">kumar-etal-2026-sentence</identifier>
<identifier type="doi">10.63317/2fe72eyjbmj3</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.61/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>818</start>
<end>827</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Sentence-Level Back-Transliteration of Romanized Indian Languages: Performance Analysis and Challenges
%A Kumar, Saurabh
%A Kakadiya, Dhruvkumar Babubhai
%A Singh, Sanasam Ranbir
%A Nandi, Sukumar
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F kumar-etal-2026-sentence
%X The widespread use of Romanized text for Indian languages, particularly on social media platforms, poses significant challenges for natural language processing due to the lack of standardized orthography and the presence of contextual ambiguities. In this study, we explore sentence-level back-transliteration for 13 Indian languages, focusing on addressing the limitations of word-level models that fail to capture contextual dependencies. We evaluate state-of-the-art models, including fine-tuned LLaMA, mT5, and Multilingual Transformer models, comparing their performance against the baseline IndicXlit model. In addition, we conduct a comprehensive error analysis to gain deeper insights into model performance. Our results demonstrate that fine-tuned LLaMA and the proposed IndiXform model, specifically designed to leverage sentence-level context, significantly outperform zero-shot LLaMA and the IndicXlit baseline. These findings provide valuable insights into handling contextual ambiguities and enhancing the accuracy of back-transliteration systems for Indian languages.
%R 10.63317/2fe72eyjbmj3
%U https://aclanthology.org/2026.lrec-1.61/
%U https://doi.org/10.63317/2fe72eyjbmj3
%P 818-827
Markdown (Informal)
[Sentence-Level Back-Transliteration of Romanized Indian Languages: Performance Analysis and Challenges](https://aclanthology.org/2026.lrec-1.61/) (Kumar et al., LREC 2026)
ACL