Signal-Routed Temperature Scaling: Low-Capacity Risk-Conditioned Calibration for Small Validation Budgets
cs.LG, cs.AI
Submitted: 2026-09-30
Updated: 2026-09-30
Terminology
Sources
- CalArena: A Large-Scale Post-Hoc Calibration Benchmark
- Smooth ECE: Principled Reliability Diagrams via Kernel Smoothing
- Quantile Adaptive Temperature Scaling for Confidence Calibration
- Bias-Reduced Uncertainty Estimation for Deep Neural Classifiers
- Sample Margin-Aware Recalibration of Temperature Scaling
- Benchmarking Neural Network Robustness to Common Corruptions and Perturbations
- Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibration
- Confidence Calibration of Classifiers with Many Classes
- Measuring Calibration in Deep Learning
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks