<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="buljan-etal-2026-context">
    <titleInfo>
        <title>Context Is (Almost) Everything: Llama-3 on Structured Output and AMR Parsing</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Maja</namePart>
        <namePart type="family">Buljan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Stephan</namePart>
        <namePart type="family">Oepen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lilja</namePart>
        <namePart type="family">Øvrelid</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>This paper evaluates the ability of an open-source LLM (Llama-3.1) to compute sentence-level semantics and encode it in formal language. We here compare two versions of the model on the task of generating a meaning representation graph for a given English sentence in the form of Abstract Meaning Representation. We explore the model’s in-context learning capability, comparing zero-shot prompting to few-shot demonstrations of varying levels of specificity. We find that Llama-3.1 frequently makes errors when reproducing the syntactic structure of both seen and unseen structured output, and that it only achieves near-SotA parsing performance when shown highly specific demonstrations similar in structure to the target sentence graph. We include an in-depth analysis of the model output, considering performance through the lens of fine-grained semantic phenomena, graph properties (e.g. top node accuracy), and graph complexity.</abstract>
    <identifier type="citekey">buljan-etal-2026-context</identifier>
    <identifier type="doi">10.63317/584xu46viahy</identifier>
    <location>
        <url>https://aclanthology.org/2026.lrec-1.915/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>11679</start>
            <end>11698</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Context Is (Almost) Everything: Llama-3 on Structured Output and AMR Parsing
%A Buljan, Maja
%A Oepen, Stephan
%A Øvrelid, Lilja
%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 buljan-etal-2026-context
%X This paper evaluates the ability of an open-source LLM (Llama-3.1) to compute sentence-level semantics and encode it in formal language. We here compare two versions of the model on the task of generating a meaning representation graph for a given English sentence in the form of Abstract Meaning Representation. We explore the model’s in-context learning capability, comparing zero-shot prompting to few-shot demonstrations of varying levels of specificity. We find that Llama-3.1 frequently makes errors when reproducing the syntactic structure of both seen and unseen structured output, and that it only achieves near-SotA parsing performance when shown highly specific demonstrations similar in structure to the target sentence graph. We include an in-depth analysis of the model output, considering performance through the lens of fine-grained semantic phenomena, graph properties (e.g. top node accuracy), and graph complexity.
%R 10.63317/584xu46viahy
%U https://aclanthology.org/2026.lrec-1.915/
%U https://doi.org/10.63317/584xu46viahy
%P 11679-11698

Markdown (Informal)

[Context Is (Almost) Everything: Llama-3 on Structured Output and AMR Parsing](https://aclanthology.org/2026.lrec-1.915/) (Buljan et al., LREC 2026)

• Context Is (Almost) Everything: Llama-3 on Structured Output and AMR Parsing (Buljan et al., LREC 2026)

ACL

• Maja Buljan, Stephan Oepen, and Lilja Øvrelid. 2026. Context Is (Almost) Everything: Llama-3 on Structured Output and AMR Parsing. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 11679–11698, Palma de Mallorca, Spain. ELRA Language Resource Association.