@inproceedings{mubarak-al-moubayed-2026-translation,
title = "Does Translation Preserve Sentiment? An Analysis of {A}rabic-{E}nglish Cross-Lingual Classification",
author = "Mubarak, Nour Aldin Al and
Al Moubayed, Noura",
editor = "Al-Khalifa, Hend and
El-Haj, Mo and
Ezzini, Saad",
booktitle = "The 7th Workshop on Open-Source {A}rabic Corpora and Processing Tools ({OSACT}7) with 5 Shared Tasks",
month = may,
year = "2026",
address = "Palma, Mallorca (Spain)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.osact-1.3/",
doi = "10.63317/2khexrk5s3bx",
pages = "25--34",
abstract = "Machine translation is widely used in cross-lingual sentiment analysis, yet the assumption that translation preserves sentiment remains largely unexamined. We present a systematic analysis of translation-induced sentiment shifts across 11,558 samples from three Arabic-English datasets (AJGT, OCLAR, FSA) using three translation models (Helsinki-NMT, GPT-4o-mini, LLaMA-3.1-8B) and a fixed multilingual classifier (XLM-RoBERTa). A substantial proportion of samples experience sentiment shifts after translation, with accuracy drops ranging from less than 1{\%} to nearly 20{\%}. GPT-4o-mini achieves the strongest sentiment preservation, while LLaMA-3.1-8B exhibits both significant distortion and refusal behaviour. Critically, Helsinki-NMT{'}s successful translation of all samples indicates that LLaMA{'}s refusals stem from safety policies rather than input untranslatability. We also find that sentiment shift measurements are pipeline-dependent and vary with the classifier used for evaluation. These findings challenge the translate-then-classify paradigm and provide guidance for cross-lingual Arabic NLP systems."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mubarak-al-moubayed-2026-translation">
<titleInfo>
<title>Does Translation Preserve Sentiment? An Analysis of Arabic-English Cross-Lingual Classification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nour</namePart>
<namePart type="given">Aldin</namePart>
<namePart type="given">Al</namePart>
<namePart type="family">Mubarak</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Noura</namePart>
<namePart type="family">Al Moubayed</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>The 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7) with 5 Shared Tasks</title>
</titleInfo>
<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">Mo</namePart>
<namePart type="family">El-Haj</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Saad</namePart>
<namePart type="family">Ezzini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Machine translation is widely used in cross-lingual sentiment analysis, yet the assumption that translation preserves sentiment remains largely unexamined. We present a systematic analysis of translation-induced sentiment shifts across 11,558 samples from three Arabic-English datasets (AJGT, OCLAR, FSA) using three translation models (Helsinki-NMT, GPT-4o-mini, LLaMA-3.1-8B) and a fixed multilingual classifier (XLM-RoBERTa). A substantial proportion of samples experience sentiment shifts after translation, with accuracy drops ranging from less than 1% to nearly 20%. GPT-4o-mini achieves the strongest sentiment preservation, while LLaMA-3.1-8B exhibits both significant distortion and refusal behaviour. Critically, Helsinki-NMT’s successful translation of all samples indicates that LLaMA’s refusals stem from safety policies rather than input untranslatability. We also find that sentiment shift measurements are pipeline-dependent and vary with the classifier used for evaluation. These findings challenge the translate-then-classify paradigm and provide guidance for cross-lingual Arabic NLP systems.</abstract>
<identifier type="citekey">mubarak-al-moubayed-2026-translation</identifier>
<identifier type="doi">10.63317/2khexrk5s3bx</identifier>
<location>
<url>https://aclanthology.org/2026.osact-1.3/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>25</start>
<end>34</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Does Translation Preserve Sentiment? An Analysis of Arabic-English Cross-Lingual Classification
%A Mubarak, Nour Aldin Al
%A Al Moubayed, Noura
%Y Al-Khalifa, Hend
%Y El-Haj, Mo
%Y Ezzini, Saad
%S The 7th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT7) with 5 Shared Tasks
%D 2026
%8 May
%I Association for Computational Linguistics
%C Palma, Mallorca (Spain)
%F mubarak-al-moubayed-2026-translation
%X Machine translation is widely used in cross-lingual sentiment analysis, yet the assumption that translation preserves sentiment remains largely unexamined. We present a systematic analysis of translation-induced sentiment shifts across 11,558 samples from three Arabic-English datasets (AJGT, OCLAR, FSA) using three translation models (Helsinki-NMT, GPT-4o-mini, LLaMA-3.1-8B) and a fixed multilingual classifier (XLM-RoBERTa). A substantial proportion of samples experience sentiment shifts after translation, with accuracy drops ranging from less than 1% to nearly 20%. GPT-4o-mini achieves the strongest sentiment preservation, while LLaMA-3.1-8B exhibits both significant distortion and refusal behaviour. Critically, Helsinki-NMT’s successful translation of all samples indicates that LLaMA’s refusals stem from safety policies rather than input untranslatability. We also find that sentiment shift measurements are pipeline-dependent and vary with the classifier used for evaluation. These findings challenge the translate-then-classify paradigm and provide guidance for cross-lingual Arabic NLP systems.
%R 10.63317/2khexrk5s3bx
%U https://aclanthology.org/2026.osact-1.3/
%U https://doi.org/10.63317/2khexrk5s3bx
%P 25-34
Markdown (Informal)
[Does Translation Preserve Sentiment? An Analysis of Arabic-English Cross-Lingual Classification](https://aclanthology.org/2026.osact-1.3/) (Mubarak & Al Moubayed, OSACT 2026)
ACL