LLMs4OL 2025: Large Language Models for Ontology Learning
The 2nd LLMs4OL Challenge @ ISWC 2025
ISWC 2025, Nara, Japan | 2-6 November
ISWC 2025, Nara, Japan | 2-6 November
Once domain-relevant terms and types are extracted (as we explored in Task A - Text2Onto), the next step is to assign a generalized type to each lexical term. This process involves mapping lexical items to their most appropriate semantic categories or ontological classes. For example, in the biomedical domain, the term “aspirin” should be classified under “Pharmaceutical Drug”. This task is crucial for organizing extracted terms into structured ontologies and improving knowledge reuse.
Given a lexical term, identify the lexical term types.
Based on the Ontology for Biomedical Investigations, focusing on biomedical terms and experimental entities.
Uses the Material Ontology to classify materials, processes, and properties in material science.
Draws from the SWEET Ontology, focusing on earth and environmental science concepts.
The datasets per given SubTasks B1, B2, and B3 are available at the "TaskB-TermTyping/" directory of the challenge repository here https://github.com/sciknoworg/LLMs4OL-Challenge/tree/main/2025.
Per subtasks, a single file was given with the following format.
[
{
'id': 'TT_e707b15a',
'term': 'unit_gigapascal',
'types': ['pressure unit']
},
...
]
⚠️ Note: In some cases may involve multiple types rather than just one.
🧪 During the testing phase, each sample will be provided with its corresponding term, and participants are expected to predict the appropriate type(s) for each sample