<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2026-nldata2opt">
    <titleInfo>
        <title>NLData2Opt: Benchmarking Parameter Derivation and Prediction for Modeling Optimization Problems from Natural Language and Data</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Lijia</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Kyohei</namePart>
        <namePart type="family">Atarashi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jiyi</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Koh</namePart>
        <namePart type="family">Takeuchi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Shunji</namePart>
        <namePart type="family">Umetani</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hisashi</namePart>
        <namePart type="family">Kashima</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>Formulating optimization problems from raw inputs of problem descriptions and data is time-consuming and costly. While Large Language Models (LLMs) offer a promising way to automate this process, existing benchmarks overestimate their capabilities by relying on idealized inputs where all data is relevant and complete. This diverges from reality, where data-centric reasoning is required to extract, compute, and predict parameters from noisy sources. To address this, we introduce NLData2Opt, a benchmark to evaluate LLMs on the multi-stage reasoning inherent in practical optimization workflows. NLData2Opt decomposes problems into problem descriptions and data, and requires models to (1) extract relevant parameters from data containing misleading attributes, (2) compute missing parameters derived by transforming existing data attributes, and (3) predict unavailable future values from historical observations. Evaluation results show substantial performance decline relative to existing benchmarks, indicating that benchmark progress does not directly translate to formulation accuracy in practical optimization workflows. NLData2Opt is available at https://github.com/sumarivyn/NLData2Opt.</abstract>
    <identifier type="citekey">liu-etal-2026-nldata2opt</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.2282/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>51908</start>
            <end>51933</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T NLData2Opt: Benchmarking Parameter Derivation and Prediction for Modeling Optimization Problems from Natural Language and Data
%A Liu, Lijia
%A Atarashi, Kyohei
%A Li, Jiyi
%A Takeuchi, Koh
%A Umetani, Shunji
%A Kashima, Hisashi
%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-nldata2opt
%X Formulating optimization problems from raw inputs of problem descriptions and data is time-consuming and costly. While Large Language Models (LLMs) offer a promising way to automate this process, existing benchmarks overestimate their capabilities by relying on idealized inputs where all data is relevant and complete. This diverges from reality, where data-centric reasoning is required to extract, compute, and predict parameters from noisy sources. To address this, we introduce NLData2Opt, a benchmark to evaluate LLMs on the multi-stage reasoning inherent in practical optimization workflows. NLData2Opt decomposes problems into problem descriptions and data, and requires models to (1) extract relevant parameters from data containing misleading attributes, (2) compute missing parameters derived by transforming existing data attributes, and (3) predict unavailable future values from historical observations. Evaluation results show substantial performance decline relative to existing benchmarks, indicating that benchmark progress does not directly translate to formulation accuracy in practical optimization workflows. NLData2Opt is available at https://github.com/sumarivyn/NLData2Opt.
%U https://aclanthology.org/2026.emnlp-main.2282/
%P 51908-51933

Markdown (Informal)

[NLData2Opt: Benchmarking Parameter Derivation and Prediction for Modeling Optimization Problems from Natural Language and Data](https://aclanthology.org/2026.emnlp-main.2282/) (Liu et al., EMNLP 2026)

• NLData2Opt: Benchmarking Parameter Derivation and Prediction for Modeling Optimization Problems from Natural Language and Data (Liu et al., EMNLP 2026)

ACL

• Lijia Liu, Kyohei Atarashi, Jiyi Li, Koh Takeuchi, Shunji Umetani, and Hisashi Kashima. 2026. NLData2Opt: Benchmarking Parameter Derivation and Prediction for Modeling Optimization Problems from Natural Language and Data. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 51908–51933, Budapest, Hungary. Association for Computational Linguistics.