<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="klemen-etal-2026-towards">
    <titleInfo>
        <title>Towards Corpus-Grounded Agentic LLMs for Multilingual Grammatical Analysis</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Matej</namePart>
        <namePart type="family">Klemen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tjaša</namePart>
        <namePart type="family">Arčon</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Luka</namePart>
        <namePart type="family">Terčon</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Marko</namePart>
        <namePart type="family">Robnik-Sikonja</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Kaja</namePart>
        <namePart type="family">Dobrovoljc</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 Workshop on Structured Linguistic Data and Evaluation (SLiDE)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Erhard</namePart>
            <namePart type="family">Hinrichs</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Joakim</namePart>
            <namePart type="family">Nivre</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Petya</namePart>
            <namePart type="family">Osenova</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">James</namePart>
            <namePart type="family">Pustejovsky</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Claus</namePart>
            <namePart type="family">Zinn</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>ELRA Language Resources Association (ELRA)</publisher>
            <place>
                <placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Empirical grammar research has become increasingly data-driven, but the systematic analysis of annotated corpora still requires substantial methodological and technical effort. We explore how agentic large language models (LLMs) can streamline this process by reasoning over annotated corpora and producing interpretable, data-grounded answers to linguistic questions. We introduce an agentic framework for corpus-grounded grammatical analysis that integrates concepts such as natural-language task interpretation, code generation, and data-driven reasoning. As a proof of concept, we apply it to Universal Dependencies (UD) corpora, testing it on multilingual grammatical tasks inspired by the World Atlas of Language Structures (WALS). The evaluation spans 13 word-order features and over 170 languages, assessing system performance across three complementary dimensions – dominant-order accuracy, order-coverage completeness, and distributional fidelity – which reflect how well the system generalizes, identifies, and quantifies word-order variations. The results demonstrate the feasibility of combining LLM reasoning with structured linguistic data, offering a first step toward interpretable, scalable automation of corpus-based grammatical inquiry.</abstract>
    <identifier type="citekey">klemen-etal-2026-towards</identifier>
    <identifier type="doi">10.63317/2vk3j6ba6zkd</identifier>
    <location>
        <url>https://aclanthology.org/2026.slide-1.12/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>136</start>
            <end>147</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Towards Corpus-Grounded Agentic LLMs for Multilingual Grammatical Analysis
%A Klemen, Matej
%A Arčon, Tjaša
%A Terčon, Luka
%A Robnik-Sikonja, Marko
%A Dobrovoljc, Kaja
%Y Hinrichs, Erhard
%Y Nivre, Joakim
%Y Osenova, Petya
%Y Pustejovsky, James
%Y Zinn, Claus
%S Proceedings of the Workshop on Structured Linguistic Data and Evaluation (SLiDE)
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma de Mallorca, Spain
%F klemen-etal-2026-towards
%X Empirical grammar research has become increasingly data-driven, but the systematic analysis of annotated corpora still requires substantial methodological and technical effort. We explore how agentic large language models (LLMs) can streamline this process by reasoning over annotated corpora and producing interpretable, data-grounded answers to linguistic questions. We introduce an agentic framework for corpus-grounded grammatical analysis that integrates concepts such as natural-language task interpretation, code generation, and data-driven reasoning. As a proof of concept, we apply it to Universal Dependencies (UD) corpora, testing it on multilingual grammatical tasks inspired by the World Atlas of Language Structures (WALS). The evaluation spans 13 word-order features and over 170 languages, assessing system performance across three complementary dimensions – dominant-order accuracy, order-coverage completeness, and distributional fidelity – which reflect how well the system generalizes, identifies, and quantifies word-order variations. The results demonstrate the feasibility of combining LLM reasoning with structured linguistic data, offering a first step toward interpretable, scalable automation of corpus-based grammatical inquiry.
%R 10.63317/2vk3j6ba6zkd
%U https://aclanthology.org/2026.slide-1.12/
%U https://doi.org/10.63317/2vk3j6ba6zkd
%P 136-147

Markdown (Informal)

[Towards Corpus-Grounded Agentic LLMs for Multilingual Grammatical Analysis](https://aclanthology.org/2026.slide-1.12/) (Klemen et al., SLiDE 2026)

• Towards Corpus-Grounded Agentic LLMs for Multilingual Grammatical Analysis (Klemen et al., SLiDE 2026)

ACL

• Matej Klemen, Tjaša Arčon, Luka Terčon, Marko Robnik-Sikonja, and Kaja Dobrovoljc. 2026. Towards Corpus-Grounded Agentic LLMs for Multilingual Grammatical Analysis. In Proceedings of the Workshop on Structured Linguistic Data and Evaluation (SLiDE), pages 136–147, Palma de Mallorca, Spain. ELRA Language Resources Association (ELRA).