@inproceedings{einarsson-2026-cross,
title = "Cross-Lingual Mathematical Reasoning in {LLM}s: Evaluating Performance on {I}celandic vs. {E}nglish Problems",
author = "Einarsson, Hafsteinn",
editor = "Morger, Felix and
Ilinykh, Nikolai and
Scalvini, Barbara and
Dobnik, Simon and
Dann{\'e}lls, Dana",
booktitle = "Proceedings of the Fourth Workshop on the Role of Resources in the Age of Large Language Models ({RESOURCEFUL} 2026)",
month = may,
year = "2026",
address = "Palma, Mallorca (Spain)",
publisher = "ELRA Language Resources Association (ELRA)",
url = "https://aclanthology.org/2026.resourceful-4.9/",
doi = "10.63317/5gybkk5wzk6g",
pages = "89--95",
abstract = "We investigate whether large language models (LLMs) exhibit performance differences when solving mathematical problems presented in a low-resource language (Icelandic) versus a high-resource language (English). Using 847 multiple-choice problems from the Icelandic Mathematics Competition corpus (STAK), we evaluate two state-of-the-art models (Gemini-3-Flash-Preview and GPT-5.4-mini) in both multiple-choice (MC) and open-ended (OE) formats, with correctness determined by a three-judge quorum (Gemini-3-Flash, GPT-5.4-mini, Claude Sonnet 4.6) achieving 97.6{\%} unanimous agreement. Our results reveal significant cross-lingual performance gaps that vary by model: Gemini-3-Flash shows a consistent English advantage of 2.4{--}10.0 percentage points across both evaluation modes, while GPT-5.4-mini exhibits no significant language effects. Notably, GPT-5.4-mini demonstrates a substantial MC deficit, achieving only 42{\%} in that format despite reaching 69-71{\%} accuracy on OE problems. Analysis of answer patterns reveals a strong option position bias in GPT-5.4-mini, with systematic over-selection of option B and under-selection of option D. These findings suggest that language does affect LLM mathematical reasoning for some models, but the effect is model-dependent and interacts with evaluation format, with implications for deploying LLMs in educational contexts for speakers of low-resource languages."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="einarsson-2026-cross">
<titleInfo>
<title>Cross-Lingual Mathematical Reasoning in LLMs: Evaluating Performance on Icelandic vs. English Problems</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hafsteinn</namePart>
<namePart type="family">Einarsson</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 Fourth Workshop on the Role of Resources in the Age of Large Language Models (RESOURCEFUL 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Felix</namePart>
<namePart type="family">Morger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikolai</namePart>
<namePart type="family">Ilinykh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="family">Scalvini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Dobnik</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dana</namePart>
<namePart type="family">Dannélls</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We investigate whether large language models (LLMs) exhibit performance differences when solving mathematical problems presented in a low-resource language (Icelandic) versus a high-resource language (English). Using 847 multiple-choice problems from the Icelandic Mathematics Competition corpus (STAK), we evaluate two state-of-the-art models (Gemini-3-Flash-Preview and GPT-5.4-mini) in both multiple-choice (MC) and open-ended (OE) formats, with correctness determined by a three-judge quorum (Gemini-3-Flash, GPT-5.4-mini, Claude Sonnet 4.6) achieving 97.6% unanimous agreement. Our results reveal significant cross-lingual performance gaps that vary by model: Gemini-3-Flash shows a consistent English advantage of 2.4–10.0 percentage points across both evaluation modes, while GPT-5.4-mini exhibits no significant language effects. Notably, GPT-5.4-mini demonstrates a substantial MC deficit, achieving only 42% in that format despite reaching 69-71% accuracy on OE problems. Analysis of answer patterns reveals a strong option position bias in GPT-5.4-mini, with systematic over-selection of option B and under-selection of option D. These findings suggest that language does affect LLM mathematical reasoning for some models, but the effect is model-dependent and interacts with evaluation format, with implications for deploying LLMs in educational contexts for speakers of low-resource languages.</abstract>
<identifier type="citekey">einarsson-2026-cross</identifier>
<identifier type="doi">10.63317/5gybkk5wzk6g</identifier>
<location>
<url>https://aclanthology.org/2026.resourceful-4.9/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>89</start>
<end>95</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Cross-Lingual Mathematical Reasoning in LLMs: Evaluating Performance on Icelandic vs. English Problems
%A Einarsson, Hafsteinn
%Y Morger, Felix
%Y Ilinykh, Nikolai
%Y Scalvini, Barbara
%Y Dobnik, Simon
%Y Dannélls, Dana
%S Proceedings of the Fourth Workshop on the Role of Resources in the Age of Large Language Models (RESOURCEFUL 2026)
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma, Mallorca (Spain)
%F einarsson-2026-cross
%X We investigate whether large language models (LLMs) exhibit performance differences when solving mathematical problems presented in a low-resource language (Icelandic) versus a high-resource language (English). Using 847 multiple-choice problems from the Icelandic Mathematics Competition corpus (STAK), we evaluate two state-of-the-art models (Gemini-3-Flash-Preview and GPT-5.4-mini) in both multiple-choice (MC) and open-ended (OE) formats, with correctness determined by a three-judge quorum (Gemini-3-Flash, GPT-5.4-mini, Claude Sonnet 4.6) achieving 97.6% unanimous agreement. Our results reveal significant cross-lingual performance gaps that vary by model: Gemini-3-Flash shows a consistent English advantage of 2.4–10.0 percentage points across both evaluation modes, while GPT-5.4-mini exhibits no significant language effects. Notably, GPT-5.4-mini demonstrates a substantial MC deficit, achieving only 42% in that format despite reaching 69-71% accuracy on OE problems. Analysis of answer patterns reveals a strong option position bias in GPT-5.4-mini, with systematic over-selection of option B and under-selection of option D. These findings suggest that language does affect LLM mathematical reasoning for some models, but the effect is model-dependent and interacts with evaluation format, with implications for deploying LLMs in educational contexts for speakers of low-resource languages.
%R 10.63317/5gybkk5wzk6g
%U https://aclanthology.org/2026.resourceful-4.9/
%U https://doi.org/10.63317/5gybkk5wzk6g
%P 89-95
Markdown (Informal)
[Cross-Lingual Mathematical Reasoning in LLMs: Evaluating Performance on Icelandic vs. English Problems](https://aclanthology.org/2026.resourceful-4.9/) (Einarsson, RESOURCEFUL 2026)
ACL