We are pleased to announce that a new paper co-authored by Associate Professor Teeradaj Racharak has been accepted for presentation at the 2nd LLMs4OL Challenge @ ISWC 2025, and will appear in the TIB Open Access Publishing proceedings. The challenge is held in conjunction with the 24th International Semantic Web Conference (ISWC 2025), one of the premier venues in the field of semantic technologies and knowledge representation.

LLMs4OL Challenge

Chavakan Yimmark and Teeradaj Racharak, T-GreC at LLMs4OL 2025 Task B: A Report on Term-Typing Task of OBI dataset using LLM with k-Nearest Neighbors, LLMs4OL 2025: The 2nd Large Language Models for Ontology Learning Challenge at the 24th ISWC, Co-located with the 24th International Semantic Web Conference, Nara, Japan, November 2-6, 2025

This paper presents a hybrid approach combining large language models (LLMs) with k-nearest neighbors (k-NN) for the term-typing task on the OBI (Ontology for Biomedical Investigations) dataset, as part of the LLMs4OL 2025 Task B. The method, named T-GreC, leverages the strengths of LLMs in contextual understanding while grounding their outputs through neighborhood-based reasoning using embedding similarity.

The work addresses a key challenge in ontology learning: enabling LLMs to assign correct semantic types to domain-specific terms. By integrating symbolic structure with neural capabilities, the proposed method offers a practical direction for applying LLMs to ontology alignment and semantic enrichment tasks in the biomedical domain.

Congratulations to the authors on this achievement!