Skip to content
conftrace_
2026 EMNLP EMNLP 2026

When More Metadata Misleads: Probing Field-Level Grounding in LLM-Generated Museum Descriptions

AbstractMetadata-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

More from EMNLP 2026

All EMNLP 2026 papers →