@inproceedings{10.1145/3805696.3846060,
author = {Tan, Mary Ann and Gesese, Genet-Asefa and Sack, Harald},
title = {BibNER-Werk: Bibliographic NER for Historical German Catalogs},
year = {2026},
isbn = {9798400725975},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3805696.3846060},
doi = {10.1145/3805696.3846060},
abstract = {General-purpose named entity recognition does not identify the title and responsible person needed to link historical German bibliographic catalog strings in the Deutsche Digitale Bibliothek (DDB) to GND1 Work authorities, since the task’s label set and centuries-spanning historical register have no counterpart in standard training corpora. This paper uses ISBD punctuation as a tiered silver labeling signal and evaluates a deterministic rule-based baseline alongside zero-shot, few-shot, and fine-tuned GLiNER-multi and fine-tuned XLM-R-large on a stratified, human-annotated gold set. The rule-based baseline beats both fine-tuned systems on untrained data; on trained data, only GLiNER-multi’s fine-tuning gains a real, confirmed precision advantage, and XLM-R-large’s never surpasses the rule.},
booktitle = {Proceedings of the 26th ACM/IEEE Joint Conference on Digital Libraries},
articleno = {8},
numpages = {8},
keywords = {named entity recognition, weak supervision, bibliographic metadata, digital libraries, entity linking, FRBR},
location = {Frisco, USA},
series = {JCDL '26}
}