DBLPLink 2.0 - An Entity Linker for the DBLP Scholarly Knowledge Graph
Abstract: In this work we present an entity linker for DBLP’s 2025 version of RDF-based Knowledge Graph. Compared to the 2022 version, DBLP now considers publication venues as a new entity type called dblp:Stream. In the earlier version of DBLPLink, we trained KG-embeddings and re-rankers on a dataset to produce entity linkings. In contrast, in this work, we develop a zero-shot entity linker using LLMs using a novel method, where we re-rank candidate entities based on the log-probabilities of the”yes”token output at the penultimate layer of the LLM.
Show BibTeX
@inproceedings{DBLP:conf/semweb/BanerjeeTU25,
author = {Debayan Banerjee and
Tilahun Abedissa Taffa and
Ricardo Usbeck},
editor = {Irene Celino and
Oktie Hassanzadeh and
Abraham Bernstein and
Natasha F. Noy and
Gong Cheng and
Shenghui Wang and
Sebasti{\'{a}}n Ferrada and
Thibaut Soulard and
Kouji Kozaki and
Hideaki Takeda and
Anna Lisa Gentile},
title = {DBLPLink 2.0 - An Entity Linker for the {DBLP} Scholarly Knowledge
Graph},
booktitle = {Joint Proceedings of Industry, Doctoral Consortium, Posters and Demos
of the 24th International Semantic Web Conference {(ISWC-C} 2025),
Nara, Japan, November 2-6, 2025},
series = {{CEUR} Workshop Proceedings},
volume = {4085},
pages = {435--440},
publisher = {CEUR-WS.org},
year = {2025},
url = {https://ceur-ws.org/Vol-4085/paper73.pdf},
timestamp = {Thu, 13 Nov 2025 16:32:27 +0100},
biburl = {https://dblp.org/rec/conf/semweb/BanerjeeTU25.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}