LATIDIA · Investigación

Modelos de cimentación tabular frente al aumento del gradiente sintonizado para el abandono de las telecomunicaciones: una evaluación consciente de las ganancias y la calibración

arXiv: 2611.12942v1Tipo de anuncio: nuevo Resumen: los modelos de cimentación tabular (TFM) predicen a partir de un contexto etiquetado en una sola pasada hacia adelante. Un punto de referencia reciente informó que superan decisivamente a los conjuntos de árboles en cu

La noticia

arXiv:2611.12942v1 Announce Type: new Abstract: Tabular foundation models (TFMs) predict from a labelled context in a single forward pass. A recent benchmark reported that they decisively outperform tree ensembles for customer churn, but it compared them with untuned gradient-boosted trees (GBDTs) and scored only ranking metrics, while a retention campaign is judged in money. We re-evaluate TabPFN-2.5, TabPFN-3.5, TabPFN-3.5-Fast and TabICL v2 against Optuna-tuned GBDTs on five public telecom churn datasets over 10 seeded splits. We separate three questions: whether a model ranks churners well, whether that ranking earns more expected maximum profit (EMP), and whether its probabilities can drive a threshold policy. With a split-aware test and Holm correction