@inproceedings{liu-etal-2025-transliterations,
title = "How Transliterations Improve Crosslingual Alignment",
author = {Liu, Yihong and
Wang, Mingyang and
Kargaran, Amir Hossein and
ImaniGooghari, Ayyoob and
Xhelili, Orgest and
Ye, Haotian and
Ma, Chunlan and
Yvon, Fran{\c{c}}ois and
Sch{\"u}tze, Hinrich},
editor = "Rambow, Owen and
Wanner, Leo and
Apidianaki, Marianna and
Al-Khalifa, Hend and
Eugenio, Barbara Di and
Schockaert, Steven",
booktitle = "Proceedings of the 31st International Conference on Computational Linguistics",
month = jan,
year = "2025",
address = "Abu Dhabi, UAE",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.coling-main.165/",
pages = "2417--2433",
abstract = "Recent studies have shown that post-aligning multilingual pretrained language models (mPLMs) using alignment objectives on both original and transliterated data can improve crosslingual alignment. This improvement further leads to better crosslingual transfer performance. However, it remains unclear how and why a better crosslingual alignment is achieved, as this technique only involves transliterations, and does not use any parallel data. This paper attempts to explicitly evaluate the crosslingual alignment and identify the key elements in transliteration-based approaches that contribute to better performance. For this, we train multiple models under varying setups for two pairs of related languages: (1) Polish and Ukrainian and (2) Hindi and Urdu. To assess alignment, we define four types of similarities based on sentence representations. Our experimental results show that adding transliterations alone improves the overall similarities, even for random sentence pairs. With the help of auxiliary transliteration-based alignment objectives, especially the contrastive objective, the model learns to distinguish matched from random pairs, leading to better crosslingual alignment. However, we also show that better alignment does not always yield better downstream performance, suggesting that further research is needed to clarify the connection between alignment and performance. The code implementation is based on \url{https://github.com/cisnlp/Transliteration-PPA}."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2025-transliterations">
<titleInfo>
<title>How Transliterations Improve Crosslingual Alignment</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yihong</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingyang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Amir</namePart>
<namePart type="given">Hossein</namePart>
<namePart type="family">Kargaran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayyoob</namePart>
<namePart type="family">ImaniGooghari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Orgest</namePart>
<namePart type="family">Xhelili</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haotian</namePart>
<namePart type="family">Ye</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chunlan</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">François</namePart>
<namePart type="family">Yvon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hinrich</namePart>
<namePart type="family">Schütze</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-01</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 31st International Conference on Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Owen</namePart>
<namePart type="family">Rambow</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leo</namePart>
<namePart type="family">Wanner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marianna</namePart>
<namePart type="family">Apidianaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hend</namePart>
<namePart type="family">Al-Khalifa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="given">Di</namePart>
<namePart type="family">Eugenio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steven</namePart>
<namePart type="family">Schockaert</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, UAE</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recent studies have shown that post-aligning multilingual pretrained language models (mPLMs) using alignment objectives on both original and transliterated data can improve crosslingual alignment. This improvement further leads to better crosslingual transfer performance. However, it remains unclear how and why a better crosslingual alignment is achieved, as this technique only involves transliterations, and does not use any parallel data. This paper attempts to explicitly evaluate the crosslingual alignment and identify the key elements in transliteration-based approaches that contribute to better performance. For this, we train multiple models under varying setups for two pairs of related languages: (1) Polish and Ukrainian and (2) Hindi and Urdu. To assess alignment, we define four types of similarities based on sentence representations. Our experimental results show that adding transliterations alone improves the overall similarities, even for random sentence pairs. With the help of auxiliary transliteration-based alignment objectives, especially the contrastive objective, the model learns to distinguish matched from random pairs, leading to better crosslingual alignment. However, we also show that better alignment does not always yield better downstream performance, suggesting that further research is needed to clarify the connection between alignment and performance. The code implementation is based on https://github.com/cisnlp/Transliteration-PPA.</abstract>
<identifier type="citekey">liu-etal-2025-transliterations</identifier>
<location>
<url>https://aclanthology.org/2025.coling-main.165/</url>
</location>
<part>
<date>2025-01</date>
<extent unit="page">
<start>2417</start>
<end>2433</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T How Transliterations Improve Crosslingual Alignment
%A Liu, Yihong
%A Wang, Mingyang
%A Kargaran, Amir Hossein
%A ImaniGooghari, Ayyoob
%A Xhelili, Orgest
%A Ye, Haotian
%A Ma, Chunlan
%A Yvon, François
%A Schütze, Hinrich
%Y Rambow, Owen
%Y Wanner, Leo
%Y Apidianaki, Marianna
%Y Al-Khalifa, Hend
%Y Eugenio, Barbara Di
%Y Schockaert, Steven
%S Proceedings of the 31st International Conference on Computational Linguistics
%D 2025
%8 January
%I Association for Computational Linguistics
%C Abu Dhabi, UAE
%F liu-etal-2025-transliterations
%X Recent studies have shown that post-aligning multilingual pretrained language models (mPLMs) using alignment objectives on both original and transliterated data can improve crosslingual alignment. This improvement further leads to better crosslingual transfer performance. However, it remains unclear how and why a better crosslingual alignment is achieved, as this technique only involves transliterations, and does not use any parallel data. This paper attempts to explicitly evaluate the crosslingual alignment and identify the key elements in transliteration-based approaches that contribute to better performance. For this, we train multiple models under varying setups for two pairs of related languages: (1) Polish and Ukrainian and (2) Hindi and Urdu. To assess alignment, we define four types of similarities based on sentence representations. Our experimental results show that adding transliterations alone improves the overall similarities, even for random sentence pairs. With the help of auxiliary transliteration-based alignment objectives, especially the contrastive objective, the model learns to distinguish matched from random pairs, leading to better crosslingual alignment. However, we also show that better alignment does not always yield better downstream performance, suggesting that further research is needed to clarify the connection between alignment and performance. The code implementation is based on https://github.com/cisnlp/Transliteration-PPA.
%U https://aclanthology.org/2025.coling-main.165/
%P 2417-2433
Markdown (Informal)
[How Transliterations Improve Crosslingual Alignment](https://aclanthology.org/2025.coling-main.165/) (Liu et al., COLING 2025)
ACL
- Yihong Liu, Mingyang Wang, Amir Hossein Kargaran, Ayyoob ImaniGooghari, Orgest Xhelili, Haotian Ye, Chunlan Ma, François Yvon, and Hinrich Schütze. 2025. How Transliterations Improve Crosslingual Alignment. In Proceedings of the 31st International Conference on Computational Linguistics, pages 2417–2433, Abu Dhabi, UAE. Association for Computational Linguistics.