@inproceedings{siro-hollink-2026-metadata,
title = "When More Metadata Misleads: Probing Field-Level Grounding in {LLM}-Generated Museum Descriptions",
author = "Siro, Clemencia and
Hollink, Laura",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.2314/",
pages = "49666--49688",
ISBN = "979-8-89176-462-0",
abstract = "Metadata-conditioned generation is often framed as a data-to-text task in which additional input fields improve grounding. Museum description generation challenges this assumption because catalogue records are heterogeneous, partially visual, and unevenly documented across cultural-geographic contexts. We study how metadata composition shapes artifact description generation through a two-phase evaluation. Phase 1 analyses natural metadata variation in a large museum collection, where field count, field type, image availability, and Place co-vary. Phase 2 introduces controlled field probes that hold object identity fixed while varying metadata quantity, field type, and image access. Our results show that metadata quantity alone does not explain generation behaviour. Field type determines whether models ground, omit, or infer information, while images can both support visual grounding and increase unsupported interpretation. More broadly, our findings show that metadata-conditioned data-to-text evaluation should assess field- and claim-level support, rather than relying only on aggregate input coverage or reference similarity"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="siro-hollink-2026-metadata">
<titleInfo>
<title>When More Metadata Misleads: Probing Field-Level Grounding in LLM-Generated Museum Descriptions</title>
</titleInfo>
<name type="personal">
<namePart type="given">Clemencia</namePart>
<namePart type="family">Siro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Laura</namePart>
<namePart type="family">Hollink</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Metadata-conditioned generation is often framed as a data-to-text task in which additional input fields improve grounding. Museum description generation challenges this assumption because catalogue records are heterogeneous, partially visual, and unevenly documented across cultural-geographic contexts. We study how metadata composition shapes artifact description generation through a two-phase evaluation. Phase 1 analyses natural metadata variation in a large museum collection, where field count, field type, image availability, and Place co-vary. Phase 2 introduces controlled field probes that hold object identity fixed while varying metadata quantity, field type, and image access. Our results show that metadata quantity alone does not explain generation behaviour. Field type determines whether models ground, omit, or infer information, while images can both support visual grounding and increase unsupported interpretation. More broadly, our findings show that metadata-conditioned data-to-text evaluation should assess field- and claim-level support, rather than relying only on aggregate input coverage or reference similarity</abstract>
<identifier type="citekey">siro-hollink-2026-metadata</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.2314/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>49666</start>
<end>49688</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T When More Metadata Misleads: Probing Field-Level Grounding in LLM-Generated Museum Descriptions
%A Siro, Clemencia
%A Hollink, Laura
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F siro-hollink-2026-metadata
%X Metadata-conditioned generation is often framed as a data-to-text task in which additional input fields improve grounding. Museum description generation challenges this assumption because catalogue records are heterogeneous, partially visual, and unevenly documented across cultural-geographic contexts. We study how metadata composition shapes artifact description generation through a two-phase evaluation. Phase 1 analyses natural metadata variation in a large museum collection, where field count, field type, image availability, and Place co-vary. Phase 2 introduces controlled field probes that hold object identity fixed while varying metadata quantity, field type, and image access. Our results show that metadata quantity alone does not explain generation behaviour. Field type determines whether models ground, omit, or infer information, while images can both support visual grounding and increase unsupported interpretation. More broadly, our findings show that metadata-conditioned data-to-text evaluation should assess field- and claim-level support, rather than relying only on aggregate input coverage or reference similarity
%U https://aclanthology.org/2026.findings-emnlp.2314/
%P 49666-49688
Markdown (Informal)
[When More Metadata Misleads: Probing Field-Level Grounding in LLM-Generated Museum Descriptions](https://aclanthology.org/2026.findings-emnlp.2314/) (Siro & Hollink, Findings 2026)
ACL