3 papers
cs.LG2026
Revisiting Explainable AI through Model-Independent Concept Dictionaries
Thomas Schnake, Doreen Schöppenthau, Alexander Meyer +3
Modern applications of AI rely on increasingly complex models. Explainable AI (XAI) has emerged as a set of techniques aimed at improving model transparency. However, existing XAI…
cs.LG2026
ReLaG: A Scalable Framework Generalizing Random Splits to Data with Latent Relations
Anthony Lavertu, Jacob Cote, Sophie Gobeil +3
Random splitting can yield non-independent train--test subsets when a dataset contains related samples, as is common in certain applications such as biochemical studies. This leads…
q-bio.QM2026
Refnd: Preventing Data Leakage in Relational Datasets
Anthony Lavertu, Jacob Cote, Jacques Corbeil +2
Machine learning models trained on biochemical data are routinely evaluated using splits that fail to account for relational structure, causing information leakage and over-optimis…