<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2026-locodec">
    <titleInfo>
        <title>LoCoDec: Robust Decoding for LLM-based Sequence-Structured Prediction</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Chang</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Peiyan</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yubo</namePart>
        <namePart type="family">Feng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Libang</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Feiyan</namePart>
        <namePart type="family">Jiang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yan</namePart>
        <namePart type="family">Song</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>Many Natural Language Processing (NLP) tasks, including information extraction, dialogue understanding, and sentiment analysis, can be formulated as structured prediction over sequences or spans. Although Large Language Model (LLM)-based methods have achieved strong performance on standard benchmarks for such tasks, their robustness under natural sentence-level perturbations and their controllability in producing valid structures remain insufficiently studied. In this paper, we propose Locality-aware Constrained Decoding (LoCoDec), a method for LLM-based structured prediction over sequences and spans that formulates decoding as constrained token paraphrasing. LoCoDec combines neighborhood context, which reduces long-text noise, with structural decoding constraints that enforce source-faithful and well-formed outputs. Experiments on sixteen benchmarks of six tasks show that LoCoDec remains competitive in S-F1 and P-F1, while achieving the best R-F1 on all datasets. Ablation results show that removing neighborhood context reduces R-F1 by up to 16.23 points. Overall, LoCoDec improves the robustness and structural consistency of LLM-based structured prediction.</abstract>
    <identifier type="citekey">liu-etal-2026-locodec</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1457/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>33312</start>
            <end>33330</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T LoCoDec: Robust Decoding for LLM-based Sequence-Structured Prediction
%A Liu, Chang
%A Wang, Peiyan
%A Feng, Yubo
%A Wang, Libang
%A Jiang, Feiyan
%A Song, Yan
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F liu-etal-2026-locodec
%X Many Natural Language Processing (NLP) tasks, including information extraction, dialogue understanding, and sentiment analysis, can be formulated as structured prediction over sequences or spans. Although Large Language Model (LLM)-based methods have achieved strong performance on standard benchmarks for such tasks, their robustness under natural sentence-level perturbations and their controllability in producing valid structures remain insufficiently studied. In this paper, we propose Locality-aware Constrained Decoding (LoCoDec), a method for LLM-based structured prediction over sequences and spans that formulates decoding as constrained token paraphrasing. LoCoDec combines neighborhood context, which reduces long-text noise, with structural decoding constraints that enforce source-faithful and well-formed outputs. Experiments on sixteen benchmarks of six tasks show that LoCoDec remains competitive in S-F1 and P-F1, while achieving the best R-F1 on all datasets. Ablation results show that removing neighborhood context reduces R-F1 by up to 16.23 points. Overall, LoCoDec improves the robustness and structural consistency of LLM-based structured prediction.
%U https://aclanthology.org/2026.emnlp-main.1457/
%P 33312-33330

Markdown (Informal)

[LoCoDec: Robust Decoding for LLM-based Sequence-Structured Prediction](https://aclanthology.org/2026.emnlp-main.1457/) (Liu et al., EMNLP 2026)

• LoCoDec: Robust Decoding for LLM-based Sequence-Structured Prediction (Liu et al., EMNLP 2026)

ACL

• Chang Liu, Peiyan Wang, Yubo Feng, Libang Wang, Feiyan Jiang, and Yan Song. 2026. LoCoDec: Robust Decoding for LLM-based Sequence-Structured Prediction. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 33312–33330, Budapest, Hungary. Association for Computational Linguistics.