AI RESEARCH
GRAFT: Decoupling Ranking and Calibration for Survival Analysis
arXiv CS.LG
•
ArXi:2602.07884v2 Announce Type: replace Survival analysis is complicated by censored data, high-dimensional features, and non-linear interactions. Classical models offer interpretability and superior calibration but are restricted to linear or predefined functional forms, while deep learning models are flexible and achieve strong discriminative performance, but tend to produce poorly calibrated survival estimates. To address this trade-off, we propose GRAFT (Gated Residual Accelerated Failure Time), a novel AFT model that decouples prognostic ranking from survival calibration.