<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="schulten-etal-2026-practitioners">
    <titleInfo>
        <title>A Practitioner’s Guide to Systematic Prompt Evaluation for Information Extraction</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Anna</namePart>
        <namePart type="given">Steinberg</namePart>
        <namePart type="family">Schulten</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Malte</namePart>
        <namePart type="family">Schierholz</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lisa</namePart>
        <namePart type="family">Reichenbach</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Frauke</namePart>
        <namePart type="family">Kreuter</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Alexander</namePart>
        <namePart type="family">Fraser</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: Industry Track</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Pepa</namePart>
            <namePart type="family">Atanasova</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Gholamreza</namePart>
            <namePart type="family">Haffari</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Minghan</namePart>
            <namePart type="family">Wang</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">979-8-89176-460-6</identifier>
    </relatedItem>
    <abstract>Practitioners across industries increasingly use large language models to extract information from unstructured documents such as financial filings, legal contracts, and sustainability reports, yet typically rely on generic online recommendations for prompt design that lack empirical grounding in their target setting. We address this gap with a practical methodology for systematically varying individual prompt components (prompt structure, contextual information, and model selection) and measuring their effect on extraction quality. We use greenhouse gas emissions extraction from corporate sustainability reports as a case study of fixed numeric extraction from high-stakes PDF documents that vary widely in structure and presentation and are dispersed across company websites. Instantiated in a retrieval-augmented generation pipeline, our evaluation shows that best practices from industry-facing prompting guides do not always hold: a structured prompt incorporating roles, definitions, extraction rules, and JSON formatting unexpectedly underperforms a simpler question-style prompt, while the benefit of adding tabular context proves both prompt- and model-specific. We condense our evaluation procedure and findings into a practitioner’s decision guide for systematically testing design choices on task-specific data rather than adopting generic recommendations by default.</abstract>
    <identifier type="citekey">schulten-etal-2026-practitioners</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-industry.59/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>975</start>
            <end>987</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T A Practitioner’s Guide to Systematic Prompt Evaluation for Information Extraction
%A Schulten, Anna Steinberg
%A Schierholz, Malte
%A Reichenbach, Lisa
%A Kreuter, Frauke
%A Fraser, Alexander
%Y Atanasova, Pepa
%Y Haffari, Gholamreza
%Y Wang, Minghan
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-460-6
%F schulten-etal-2026-practitioners
%X Practitioners across industries increasingly use large language models to extract information from unstructured documents such as financial filings, legal contracts, and sustainability reports, yet typically rely on generic online recommendations for prompt design that lack empirical grounding in their target setting. We address this gap with a practical methodology for systematically varying individual prompt components (prompt structure, contextual information, and model selection) and measuring their effect on extraction quality. We use greenhouse gas emissions extraction from corporate sustainability reports as a case study of fixed numeric extraction from high-stakes PDF documents that vary widely in structure and presentation and are dispersed across company websites. Instantiated in a retrieval-augmented generation pipeline, our evaluation shows that best practices from industry-facing prompting guides do not always hold: a structured prompt incorporating roles, definitions, extraction rules, and JSON formatting unexpectedly underperforms a simpler question-style prompt, while the benefit of adding tabular context proves both prompt- and model-specific. We condense our evaluation procedure and findings into a practitioner’s decision guide for systematically testing design choices on task-specific data rather than adopting generic recommendations by default.
%U https://aclanthology.org/2026.emnlp-industry.59/
%P 975-987

Markdown (Informal)

[A Practitioner’s Guide to Systematic Prompt Evaluation for Information Extraction](https://aclanthology.org/2026.emnlp-industry.59/) (Schulten et al., EMNLP 2026)

• A Practitioner’s Guide to Systematic Prompt Evaluation for Information Extraction (Schulten et al., EMNLP 2026)

ACL

• Anna Steinberg Schulten, Malte Schierholz, Lisa Reichenbach, Frauke Kreuter, and Alexander Fraser. 2026. A Practitioner’s Guide to Systematic Prompt Evaluation for Information Extraction. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 975–987, Budapest, Hungary. Association for Computational Linguistics.