Comprehensive Categorisation of AI Evaluation Methods

Teyavongsak, Jom · Zenodo (CERN European Organization for Nuclear Research) · 2025

Systematising the diverse landscape of AI evaluation into seven paradigms and respective sub-paradigms to offer clear distinctions and clarify the specific purpose and use case of each AI Evaluation method. Hence, enabling researchers and practitioners to better select suitable approaches and understand the implications of evaluation outcomes.

Read the paper · More papers on PaperTik