Ontology Alignment Goes Wide
Year: 2026
Type: article
Abstract: We have made significant progress in multiple tasks related to ontology alignment with much of that progress thanks to the work of the rigorous evaluation environments of the OAEI. The progress has only accelerated thanks to the capabilities of LLMs. In this talk, I first discuss our recent work on how to efficiently use LLMs for ontology alignment related tasks specifically entity matching (i.e. instance matching) and entity linking. I use this work to motivate the idea that ontology alignment is a critical task with respect to both agentic memory and cross-agent collaboration. Given alignments important role, I argue that our evaluation environments need to expand to consider broader notions of ontology and data sources going beyond what is currently considered in the OAEI. Keynote presentation at the The 21st International Workshop on Ontology Matching (OM 2026) collocated with the 25th International Semantic Web Conference ISWC-2026 October 25th, 2026, Bari, Italy
Author: Groth, Paul
Institution: University of Amsterdam
Language: English
Topic: Semantic Web and Ontologies
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7223552463
