Tag
This paper proposes CORD, a post-fit adapter for post-hoc calibration that repairs probability vectors to preserve original top-1 predictions while maintaining calibration quality, achieving zero TPCR and improved metrics on datasets like CIFAR and ImageNet.
The paper proposes Invertible Logits Transformation (InvLT), a post-hoc calibration method that applies a learnable scalar MLP element-wise to logits, preserving accuracy while correcting nonlinear miscalibration more flexibly than temperature scaling.