<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="einarsson-2026-mazeeval">
    <titleInfo>
        <title>MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models</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 Fifteenth Language Resources and Evaluation Conference</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Stelios</namePart>
            <namePart type="family">Piperidis</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Núria</namePart>
            <namePart type="family">Bel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Henk</namePart>
            <namePart type="family">van den Heuvel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nancy</namePart>
            <namePart type="family">Ide</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Simon</namePart>
            <namePart type="family">Krek</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Antonio</namePart>
            <namePart type="family">Toral</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>ELRA Language Resource Association</publisher>
            <place>
                <placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>As Large Language Models (LLMs) increasingly power autonomous agents in robotics and embodied AI, understanding their spatial reasoning capabilities becomes crucial for reliable deployment. We introduce MazeEval, a benchmark designed to evaluate pure spatial reasoning in LLMs through coordinate-based maze navigation tasks without visual input. Using a function-calling interface, models navigate mazes of varying complexity (5 x 5 to 15 x 15 grids) using only coordinate feedback and distance-to-wall information. We evaluate eight state-of-the-art LLMs across identical mazes in both English and Icelandic to assess cross-linguistic transfer of spatial abilities. Our findings reveal striking disparities: while OpenAI’s O3 achieves perfect navigation up to 30 x 30 mazes, other models exhibit catastrophic failure beyond 9 x 9 mazes, with 100% of failures attributed to excessive looping behavior. We document significant performance degradation in Icelandic, with models solving mazes 3-4 sizes smaller than in English, suggesting spatial reasoning emerges from linguistic patterns rather than language-agnostic mechanisms. These results highlight that spatial intelligence remains fundamentally constrained by training data availability, with important implications for global deployment of LLM-powered autonomous systems.</abstract>
    <identifier type="citekey">einarsson-2026-mazeeval</identifier>
    <identifier type="doi">10.63317/4nm93hckcaf2</identifier>
    <location>
        <url>https://aclanthology.org/2026.lrec-1.27/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>407</start>
            <end>418</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models
%A Einarsson, Hafsteinn
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F einarsson-2026-mazeeval
%X As Large Language Models (LLMs) increasingly power autonomous agents in robotics and embodied AI, understanding their spatial reasoning capabilities becomes crucial for reliable deployment. We introduce MazeEval, a benchmark designed to evaluate pure spatial reasoning in LLMs through coordinate-based maze navigation tasks without visual input. Using a function-calling interface, models navigate mazes of varying complexity (5 x 5 to 15 x 15 grids) using only coordinate feedback and distance-to-wall information. We evaluate eight state-of-the-art LLMs across identical mazes in both English and Icelandic to assess cross-linguistic transfer of spatial abilities. Our findings reveal striking disparities: while OpenAI’s O3 achieves perfect navigation up to 30 x 30 mazes, other models exhibit catastrophic failure beyond 9 x 9 mazes, with 100% of failures attributed to excessive looping behavior. We document significant performance degradation in Icelandic, with models solving mazes 3-4 sizes smaller than in English, suggesting spatial reasoning emerges from linguistic patterns rather than language-agnostic mechanisms. These results highlight that spatial intelligence remains fundamentally constrained by training data availability, with important implications for global deployment of LLM-powered autonomous systems.
%R 10.63317/4nm93hckcaf2
%U https://aclanthology.org/2026.lrec-1.27/
%U https://doi.org/10.63317/4nm93hckcaf2
%P 407-418

Markdown (Informal)

[MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models](https://aclanthology.org/2026.lrec-1.27/) (Einarsson, LREC 2026)

• MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models (Einarsson, LREC 2026)

ACL

• Hafsteinn Einarsson. 2026. MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 407–418, Palma de Mallorca, Spain. ELRA Language Resource Association.