Cross-Lingual Mathematical Reasoning in LLMs: Evaluating Performance on Icelandic vs. English Problems

Hafsteinn Einarsson


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.
Anthology ID:
2026.resourceful-4.9
Volume:
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)
Editors:
Felix Morger, Nikolai Ilinykh, Barbara Scalvini, Simon Dobnik, Dana Dannélls
Venues:
RESOURCEFUL | WS
SIG:
Publisher:
ELRA Language Resources Association (ELRA)
Note:
Pages:
89–95
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-resourceful-09
DOI:
10.63317/5gybkk5wzk6g
Bibkey:
Cite (ACL):
Hafsteinn Einarsson. 2026. Cross-Lingual Mathematical Reasoning in LLMs: Evaluating Performance on Icelandic vs. English Problems. In Proceedings of the Fourth Workshop on the Role of Resources in the Age of Large Language Models (RESOURCEFUL 2026), pages 89–95, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).
Cite (Informal):
Cross-Lingual Mathematical Reasoning in LLMs: Evaluating Performance on Icelandic vs. English Problems (Einarsson, RESOURCEFUL 2026)
Copy Citation: