mlsanityData to 14 Oct 2026

Post-Hoc Split-Point Self-Consistency Verification for Efficient, Unified Quantification of Aleatoric and Epistemic Uncertainty in Deep Learning

arXiv 2509.13262 ↗First posted 16 Sep 2025cs.LGcited by 0

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Uncertainty quantification (UQ) is vital for trustworthy deep learning, yet existing methods are either computationally intensive, such as Bayesian or ensemble methods, or provide only partial, task-specific estimates, such as single-forward-pass techniques. In this paper, we propose a post-hoc single-forward-pass framework that jointly captures aleatoric and epistemic uncertainty without modifying or retraining pretrained models. Our method applies Split-Point Analysis (SPA) to decompose predictive residuals into upper and lower subsets, computing Mean Absolute Residuals (MARs) on each side. We prove that, under ideal conditions, the total MAR equals the harmonic mean of subset MARs; deviations define a novel Self-consistency Discrepancy Score (SDS) for fine-grained epistemic estimation across regression and classification. For regression, side-specific quantile regression yields prediction intervals with improved empirical coverage, which are further calibrated via SDS. For classification, when calibration data are available, we apply SPA-based calibration identities to adjust the softmax outputs and then compute predictive entropy on these calibrated probabilities. Extensive experiments on diverse regression and classification benchmarks demonstrate that our framework matches or exceeds several state-of-the-art UQ methods while incurring minimal overhead. Our source code is available at https://github.com/zzz0527/SPC-UQ.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 38 of its 68 references that are in the map. First posted on arXiv 16 Sep 2025.

Builds on

38 of 68 references.

References · 38

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  5. 2017On Calibration of Modern Neural Networkscited by 4,960
  6. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  7. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  8. 2016Wide Residual Networkscited by 3,070
  9. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,877
  10. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,352
  11. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 913
  12. 2020A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challengescited by 819
26 more
  1. 2018Evidential Deep Learning to Quantify Classification Uncertaintycited by 732
  2. 2021A Survey of Uncertainty in Deep Neural Networkscited by 680
  3. 2019A Simple Baseline for Bayesian Uncertainty in Deep Learningcited by 560
  4. 2018Predictive Uncertainty Estimation via Prior Networkscited by 557
  5. 2016Deep Exploration via Bootstrapped DQNcited by 527
  6. 2020Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awarenesscited by 389
  7. 2021Laplace Redux -- Effortless Bayesian Deep Learningcited by 365
  8. 2019Deep Evidential Regressioncited by 355
  9. 2015Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networkscited by 295
  10. 2018Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networkscited by 280
  11. 2021Deep Deterministic Uncertainty: A Simple Baselinecited by 200
  12. 2017Multiplicative Normalizing Flows for Variational Bayesian Neural Networkscited by 157
  13. 2020Uncertainty as a Form of Transparency: Measuring, Communicating, and Using Uncertaintycited by 129
  14. 2021DEUP: Direct Epistemic Uncertainty Predictioncited by 104
  15. 2019Evaluating and Calibrating Uncertainty Prediction in Regression Taskscited by 87
  16. 2022A Deeper Look into Aleatoric and Epistemic Uncertainty Disentanglementcited by 87
  17. 2020Beyond Pinball Loss: Quantile Methods for Calibrated Uncertainty Quantificationcited by 78
  18. 2022Pitfalls of Epistemic Uncertainty Quantification through Loss Minimisationcited by 66
  19. 2024Reliable Conflictive Multi-View Learningcited by 45
  20. 2022A Review of Uncertainty for Deep Reinforcement Learningcited by 40
  21. 2023On Second-Order Scoring Rules for Epistemic Uncertainty Quantificationcited by 38
  22. 2023Quantification of Uncertainty with Adversarial Modelscited by 28
  23. 2024Are Uncertainty Quantification Capabilities of Evidential Deep Learning a Mirage?cited by 18
  24. 2020Prediction Intervals: Split Normal Mixture from Quality-Driven Deep Ensemblescited by 14
  25. 2023Deep Evidential Learning for Bayesian Quantile Regressioncited by 5
  26. 2024LUMA: A Benchmark Dataset for Learning from Uncertain and Multimodal Datacited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Complementing Self-Consistency with Cross-Model Disagreement for Uncertainty Quantification
  2. 2024Towards Modeling Uncertainties of Self-explaining Neural Networks via Conformal Prediction
  3. 2024Estimating Epistemic and Aleatoric Uncertainty with a Single Model
  4. 2024Echoes of Socratic Doubt: Embracing Uncertainty in Calibrated Evidential Reinforcement Learning
  5. 2026Efficient Epistemic Uncertainty Estimation for Large Language Models via Knowledge Distillation
  6. 2026Epistemic Uncertainty Quantification for Pre-trained VLMs via Riemannian Flow Matching
  7. 2026Certified Per-Instance Unlearning Using Individual Sensitivity Bounds
  8. 2026Revealing Epistemic Uncertainty in MLLMs via Causal-Invariant Masking