Towards Accurate and Calibrated Classification: Regularizing Cross-Entropy From A Generative Perspective

ArXi:2604.06689v1 Announce Type: new Accurate classification requires not only high predictive accuracy but also well-calibrated confidence estimates. Yet, modern deep neural networks (DNNs) are often overconfident, primarily due to overfitting on the negative log-likelihood (NLL). While focal loss variants alleviate this issue, they typically reduce accuracy, revealing a persistent trade-off between calibration and predictive performance.