@inproceedings{thomas-vince-2026-1114,
title = "What 1,114 Documents Reveal About Schema-Driven Extraction: Models, Pipelines, and the Real-vs-Synthetic Gap",
author = "Thomas, Frank and
Vince, Andrei",
editor = "Atanasova, Pepa and
Haffari, Gholamreza and
Wang, Minghan",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-industry.56/",
pages = "928--940",
ISBN = "979-8-89176-460-6",
abstract = "Extracting structured data from documents is a core task in finance, insurance, legal, and healthcare, yet empirical studies comparing extraction approaches across domains remain scarce. We assemble a corpus of 1,114 documents across 12 categories spanning 8 domains, with 7,197 fields annotated with ground-truth values, and use it to study how extraction accuracy varies across models, pipeline architectures, and document sources. We evaluate with a type-aware methodology that scores every field as a correct extraction, a miss, a hallucination, or a correct absence, so that declining to extract is distinguished from extracting wrongly. Our central finding is that aggregate accuracy conceals critical divergence along every axis: synthetic documents overestimate accuracy by 38{--}45pp on insurance certificates (n=18 real documents); three models score within 1pp overall but diverge by 31pp per category; and two pipeline architectures diverge by 26pp on contracts{---}with the advantage not predicted by document length. An independently built extraction system, run under a protocol fixed in advance, agrees with our ground truth on 82{--}87{\%} of fields across three models spanning two providers, providing evidence that the corpus is not calibrated to a single implementation. The corpus and evaluation tooling are released to enable reproducible comparison."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="thomas-vince-2026-1114">
<titleInfo>
<title>What 1,114 Documents Reveal About Schema-Driven Extraction: Models, Pipelines, and the Real-vs-Synthetic Gap</title>
</titleInfo>
<name type="personal">
<namePart type="given">Frank</namePart>
<namePart type="family">Thomas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrei</namePart>
<namePart type="family">Vince</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>Extracting structured data from documents is a core task in finance, insurance, legal, and healthcare, yet empirical studies comparing extraction approaches across domains remain scarce. We assemble a corpus of 1,114 documents across 12 categories spanning 8 domains, with 7,197 fields annotated with ground-truth values, and use it to study how extraction accuracy varies across models, pipeline architectures, and document sources. We evaluate with a type-aware methodology that scores every field as a correct extraction, a miss, a hallucination, or a correct absence, so that declining to extract is distinguished from extracting wrongly. Our central finding is that aggregate accuracy conceals critical divergence along every axis: synthetic documents overestimate accuracy by 38–45pp on insurance certificates (n=18 real documents); three models score within 1pp overall but diverge by 31pp per category; and two pipeline architectures diverge by 26pp on contracts—with the advantage not predicted by document length. An independently built extraction system, run under a protocol fixed in advance, agrees with our ground truth on 82–87% of fields across three models spanning two providers, providing evidence that the corpus is not calibrated to a single implementation. The corpus and evaluation tooling are released to enable reproducible comparison.</abstract>
<identifier type="citekey">thomas-vince-2026-1114</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-industry.56/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>928</start>
<end>940</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T What 1,114 Documents Reveal About Schema-Driven Extraction: Models, Pipelines, and the Real-vs-Synthetic Gap
%A Thomas, Frank
%A Vince, Andrei
%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 thomas-vince-2026-1114
%X Extracting structured data from documents is a core task in finance, insurance, legal, and healthcare, yet empirical studies comparing extraction approaches across domains remain scarce. We assemble a corpus of 1,114 documents across 12 categories spanning 8 domains, with 7,197 fields annotated with ground-truth values, and use it to study how extraction accuracy varies across models, pipeline architectures, and document sources. We evaluate with a type-aware methodology that scores every field as a correct extraction, a miss, a hallucination, or a correct absence, so that declining to extract is distinguished from extracting wrongly. Our central finding is that aggregate accuracy conceals critical divergence along every axis: synthetic documents overestimate accuracy by 38–45pp on insurance certificates (n=18 real documents); three models score within 1pp overall but diverge by 31pp per category; and two pipeline architectures diverge by 26pp on contracts—with the advantage not predicted by document length. An independently built extraction system, run under a protocol fixed in advance, agrees with our ground truth on 82–87% of fields across three models spanning two providers, providing evidence that the corpus is not calibrated to a single implementation. The corpus and evaluation tooling are released to enable reproducible comparison.
%U https://aclanthology.org/2026.emnlp-industry.56/
%P 928-940
Markdown (Informal)
[What 1,114 Documents Reveal About Schema-Driven Extraction: Models, Pipelines, and the Real-vs-Synthetic Gap](https://aclanthology.org/2026.emnlp-industry.56/) (Thomas & Vince, EMNLP 2026)
ACL