Jun 2026
FedLAS: Feature-Modulated Bidirectional Label Smoothing for Neural Network Calibration
This work proposes FedLAS: Feature-Modulated Bidirectional Label Smoothing, a plug-and-play algorithm for label smoothing-based losses that consistently improves calibration compared to modern baselines, reducing Expected Calibration Error (ECE) and Adaptive ECE while maintaining Top-1 accuracy.
Thiru Thillai Nadarasar Bahavan, Sachith Seneviratne, Saman K. Halgamuge
· arXiv.org · 0 citations