Ontology-Based Contextual AI Evaluations (OB-CAIE) methodology published on arXiv
A new methodology called Ontology-Based Contextual AI Evaluations (OB-CAIE) is described in arXiv:2610.00529v1; it formalizes AI evaluation problem spaces using two ontologies — a Domain-Specific Ontology (DSO) for the 'what' and an Evaluation Process Ontology (EPO) for the 'how' — to improve clarity, reproducibility, and traceability of failure points, while permitting structured human judgment where needed.