mlsanityData to 14 Oct 2026

VOLTA: The Surprising Ineffectiveness of Auxiliary Losses for Calibrated Deep Learning

arXiv 2604.08639 ↗First posted 9 Apr 2026cs.LG cs.AI cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Uncertainty quantification (UQ) is essential for deploying deep learning models in safety critical applications, yet no consensus exists on which UQ method performs best across different data modalities and distribution shifts. This paper presents a comprehensive benchmark of ten widely used UQ baselines including MC Dropout, SWAG, ensemble methods, temperature scaling, energy based OOD, Mahalanobis, hyperbolic classifiers, ENN, Taylor Sensus, and split conformal prediction against a simplified yet highly effective variant of VOLTA that retains only a deep encoder, learnable prototypes, cross entropy loss, and post hoc temperature scaling. We evaluate all methods on CIFAR 10 (in distribution), CIFAR 100, SVHN, uniform noise (out of distribution), CIFAR 10 C (corruptions), and Tiny ImageNet features (tabular). VOLTA achieves competitive or superior accuracy (up to 0.864 on CIFAR 10), significantly lower expected calibration error (0.010 vs. 0.044 to 0.102 for baselines), and strong OOD detection (AUROC 0.802). Statistical testing over three random seeds shows that VOLTA matches or outperforms most baselines, with ablation studies confirming the importance of adaptive temperature and deep encoders. Our results establish VOLTA as a lightweight, deterministic, and well calibrated alternative to more complex UQ approaches.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 41 of its 69 references that are in the map. First posted on arXiv 9 Apr 2026.

Builds on

41 of 69 references.

References · 41

  1. 2017On Calibration of Modern Neural Networkscited by 4,960
  2. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  3. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,877
  4. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
  5. 2020Energy-based Out-of-distribution Detectioncited by 1,299
  6. 2019Scaling Out-of-Distribution Detection for Real-World Settingscited by 465
  7. 2021Revisiting the Calibration of Modern Neural Networkscited by 390
  8. 2021Laplace Redux -- Effortless Bayesian Deep Learningcited by 365
  9. 2019Verified Uncertainty Calibrationcited by 309
  10. 2022OpenOOD: Benchmarking Generalized Out-of-Distribution Detectioncited by 268
  11. 2023OpenOOD v1.5: Enhanced Benchmark for Out-of-Distribution Detectioncited by 179
  12. 2019Evaluating model calibration in classificationcited by 168
29 more
  1. 2020Calibration of Neural Networks using Splinescited by 97
  2. 2020Masksembles for Uncertainty Estimationcited by 88
  3. 2022A Deeper Look into Aleatoric and Epistemic Uncertainty Disentanglementcited by 87
  4. 2023A Survey on Uncertainty Quantification Methods for Deep Learningcited by 81
  5. 2023Calibration in Deep Learning: A Survey of the State-of-the-Artcited by 75
  6. 2018Improving Reconstruction Autoencoder Out-of-distribution Detection with Mahalanobis Distancecited by 74
  7. 2021Understanding Softmax Confidence and Uncertaintycited by 63
  8. 2024Estimating Epistemic and Aleatoric Uncertainty with a Single Modelcited by 54
  9. 2022Assaying Out-Of-Distribution Generalization in Transfer Learningcited by 53
  10. 2021How to Certify Machine Learning Based Safety-critical Systems? A Systematic Literature Reviewcited by 39
  11. 2020Out-of-Distribution Detection for Automotive Perceptioncited by 37
  12. 2021Confidence-based Out-of-Distribution Detection: A Comparative Study and Analysiscited by 34
  13. 2022Metrics of calibration for probabilistic predictionscited by 32
  14. 2020Why is the Mahalanobis Distance Effective for Anomaly Detection?cited by 28
  15. 2022Accelerated Linearized Laplace Approximation for Bayesian Deep Learningcited by 28
  16. 2022Augmenting Softmax Information for Selective Classification with Out-of-Distribution Datacited by 27
  17. 2023A Large-Scale Study of Probabilistic Calibration in Neural Network Regressioncited by 27
  18. 2018Deep Ensemble Bayesian Active Learning : Addressing the Mode Collapse issue in Monte Carlo dropout via Ensemblescited by 20
  19. 2019Confidence Calibration for Convolutional Neural Networks Using Structured Dropoutcited by 19
  20. 2023Variational Linearized Laplace Approximation for Bayesian Deep Learningcited by 18
  21. 2023Riemannian Laplace approximations for Bayesian neural networkscited by 15
  22. 2023Density-Softmax: Efficient Test-time Model for Uncertainty Estimation and Robustness under Distribution Shiftscited by 12
  23. 2024A Structured Review of Literature on Uncertainty in Machine Learning & Deep Learningcited by 10
  24. 2022Raising the Bar on the Evaluation of Out-of-Distribution Detectioncited by 7
  25. 2019Are Out-of-Distribution Detection Methods Effective on Large-Scale Datasets?cited by 6
  26. 2025laplax -- Laplace Approximations with JAXcited by 5
  27. 2021L2M: Practical posterior Laplace approximation with optimization-driven second moment estimationcited by 2
  28. 2025A comprehensive review of classifier probability calibration metricscited by 2
  29. 2026Scalable Linearized Laplace Approximation via Surrogate Neural Kernelcited by 2

Built on by

Not cited yet.