SAMix: Calibrated and Accurate Continual Learning via Sphere-Adaptive Mixup and Neural Collapse

ArXi:2510.15751v2 Announce Type: replace While most continual learning methods focus on mitigating forgetting and improving accuracy, they often overlook the critical aspect of network calibration, despite its importance. Neural collapse, a phenomenon where last-layer features collapse to their class means, has nstrated advantages in continual learning by reducing feature-classifier misalignment. Few works aim to improve the calibration of continual models for reliable predictions.