ResearchBenchmarksClassification on 30 datasets aggregate (mean across 5 folds)Follow0Log-lossXGBoost-1.126.441421.56May 5, 2026Evaluation ResultsMethodMethodLinksLog-lossBrier ScoreAUC-ROCXGBoostAggregation=Head-to-he...Aggregation=Head-to-head wins, Validation=5-fold cross-validation2026.05012LightGBMAggregation=Head-to-he...Aggregation=Head-to-head wins, Validation=5-fold cross-validation2026.05234CatBoostAggregation=Head-to-he...Aggregation=Head-to-head wins, Validation=5-fold cross-validation2026.05282624