@inproceedings{meng-etal-2026-language,
title = "Do Language Models Reason Across Languages?",
author = "Meng, Yan and
Mohammed, Wafaa and
Monz, Christof",
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.2441/",
pages = "55396--55411",
ISBN = "979-8-89176-459-0",
abstract = "Real-world data sources are inherently multilingual, which raises the question whether language models can synthesize information across languages. In this paper, we introduce a simple two-hop question answering setting, where answering a question requires making inferences over two documents in different languages. We find that language models are more sensitive to language variation in answer-span documents than in those providing bridging information, despite the equal importance of both documents for answering a question. Under a step-by-step sub-question evaluation, we further show that in up to 33{\%} of multilingual cases, models fail to infer the bridging information in the first step yet still answer the overall question correctly. This indicates that reasoning in language models, especially in multilingual settings, does not follow a faithful step-by-step decomposition. Moreover, language models also struggle to compose intermediate results: in up to 18{\%} of cases, they answer both sub-questions correctly but fail to derive the correct final answer. To examine whether explicit decomposition can help, we evaluate a three-stage SUBQ prompting setup that provides the relevant intermediate sub-questions during inference. This improves two-hop QA performance on both the compositional failure cases and the full evaluation set."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="meng-etal-2026-language">
<titleInfo>
<title>Do Language Models Reason Across Languages?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yan</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wafaa</namePart>
<namePart type="family">Mohammed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christof</namePart>
<namePart type="family">Monz</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>Real-world data sources are inherently multilingual, which raises the question whether language models can synthesize information across languages. In this paper, we introduce a simple two-hop question answering setting, where answering a question requires making inferences over two documents in different languages. We find that language models are more sensitive to language variation in answer-span documents than in those providing bridging information, despite the equal importance of both documents for answering a question. Under a step-by-step sub-question evaluation, we further show that in up to 33% of multilingual cases, models fail to infer the bridging information in the first step yet still answer the overall question correctly. This indicates that reasoning in language models, especially in multilingual settings, does not follow a faithful step-by-step decomposition. Moreover, language models also struggle to compose intermediate results: in up to 18% of cases, they answer both sub-questions correctly but fail to derive the correct final answer. To examine whether explicit decomposition can help, we evaluate a three-stage SUBQ prompting setup that provides the relevant intermediate sub-questions during inference. This improves two-hop QA performance on both the compositional failure cases and the full evaluation set.</abstract>
<identifier type="citekey">meng-etal-2026-language</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2441/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>55396</start>
<end>55411</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Do Language Models Reason Across Languages?
%A Meng, Yan
%A Mohammed, Wafaa
%A Monz, Christof
%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 meng-etal-2026-language
%X Real-world data sources are inherently multilingual, which raises the question whether language models can synthesize information across languages. In this paper, we introduce a simple two-hop question answering setting, where answering a question requires making inferences over two documents in different languages. We find that language models are more sensitive to language variation in answer-span documents than in those providing bridging information, despite the equal importance of both documents for answering a question. Under a step-by-step sub-question evaluation, we further show that in up to 33% of multilingual cases, models fail to infer the bridging information in the first step yet still answer the overall question correctly. This indicates that reasoning in language models, especially in multilingual settings, does not follow a faithful step-by-step decomposition. Moreover, language models also struggle to compose intermediate results: in up to 18% of cases, they answer both sub-questions correctly but fail to derive the correct final answer. To examine whether explicit decomposition can help, we evaluate a three-stage SUBQ prompting setup that provides the relevant intermediate sub-questions during inference. This improves two-hop QA performance on both the compositional failure cases and the full evaluation set.
%U https://aclanthology.org/2026.emnlp-main.2441/
%P 55396-55411
Markdown (Informal)
[Do Language Models Reason Across Languages?](https://aclanthology.org/2026.emnlp-main.2441/) (Meng et al., EMNLP 2026)
ACL
- Yan Meng, Wafaa Mohammed, and Christof Monz. 2026. Do Language Models Reason Across Languages?. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 55396–55411, Budapest, Hungary. Association for Computational Linguistics.