@inproceedings{al-ali-etal-2026-predicting,
title = "Predicting Multilingual Classification and Translation Performance of {LLM}s with Cross-Lingual Alignment {--} Is {E}nglish Enough?",
author = {Al Ali, Adnan and
H{\"a}mmerl, Kathy and
Libovick{\'y}, Jind{\v{r}}ich and
Fraser, Alexander},
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.162/",
pages = "3719--3741",
ISBN = "979-8-89176-459-0",
abstract = "Multilingual large language models (LLMs) have been shown to perform better on non-English classification tasks when the representations of the given language are more aligned to English within the model. Several cross-lingual alignment (CLA) scores have been proposed for use with LLMs, along with multiple approaches for extracting embeddings from the models. We provide a comparative analysis of 27 CLA score variants, examining how they differ and how well each predicts downstream performance across three tasks. Crucially, while LLMs are widely used for generative tasks such as machine translation, prior work has focused almost exclusively on classification. We therefore investigate whether CLA scores are similarly predictive of translation performance. To enable computing correlations across target languages, we propose a PMI-based translation metric, which is less dependent on the target language and correlates strongly with CHRF. We find that CLA with English predicts translation quality comparably to or better than source{--}target CLA, providing new evidence that LLMs use English as an internal pivot language."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="al-ali-etal-2026-predicting">
<titleInfo>
<title>Predicting Multilingual Classification and Translation Performance of LLMs with Cross-Lingual Alignment – Is English Enough?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Adnan</namePart>
<namePart type="family">Al Ali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kathy</namePart>
<namePart type="family">Hämmerl</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jindřich</namePart>
<namePart type="family">Libovický</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Fraser</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Multilingual large language models (LLMs) have been shown to perform better on non-English classification tasks when the representations of the given language are more aligned to English within the model. Several cross-lingual alignment (CLA) scores have been proposed for use with LLMs, along with multiple approaches for extracting embeddings from the models. We provide a comparative analysis of 27 CLA score variants, examining how they differ and how well each predicts downstream performance across three tasks. Crucially, while LLMs are widely used for generative tasks such as machine translation, prior work has focused almost exclusively on classification. We therefore investigate whether CLA scores are similarly predictive of translation performance. To enable computing correlations across target languages, we propose a PMI-based translation metric, which is less dependent on the target language and correlates strongly with CHRF. We find that CLA with English predicts translation quality comparably to or better than source–target CLA, providing new evidence that LLMs use English as an internal pivot language.</abstract>
<identifier type="citekey">al-ali-etal-2026-predicting</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.162/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>3719</start>
<end>3741</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Predicting Multilingual Classification and Translation Performance of LLMs with Cross-Lingual Alignment – Is English Enough?
%A Al Ali, Adnan
%A Hämmerl, Kathy
%A Libovický, Jindřich
%A Fraser, Alexander
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F al-ali-etal-2026-predicting
%X Multilingual large language models (LLMs) have been shown to perform better on non-English classification tasks when the representations of the given language are more aligned to English within the model. Several cross-lingual alignment (CLA) scores have been proposed for use with LLMs, along with multiple approaches for extracting embeddings from the models. We provide a comparative analysis of 27 CLA score variants, examining how they differ and how well each predicts downstream performance across three tasks. Crucially, while LLMs are widely used for generative tasks such as machine translation, prior work has focused almost exclusively on classification. We therefore investigate whether CLA scores are similarly predictive of translation performance. To enable computing correlations across target languages, we propose a PMI-based translation metric, which is less dependent on the target language and correlates strongly with CHRF. We find that CLA with English predicts translation quality comparably to or better than source–target CLA, providing new evidence that LLMs use English as an internal pivot language.
%U https://aclanthology.org/2026.emnlp-main.162/
%P 3719-3741
Markdown (Informal)
[Predicting Multilingual Classification and Translation Performance of LLMs with Cross-Lingual Alignment – Is English Enough?](https://aclanthology.org/2026.emnlp-main.162/) (Al Ali et al., EMNLP 2026)
ACL