mlsanityData to 7 Oct 2026

Density of States Estimation for Out-of-Distribution Detection

arXiv 2006.09273 ↗First posted 16 Jun 2020cs.LG stat.MLcited by 74

Not on the map. Why?

Abstract (arXiv)

Perhaps surprisingly, recent studies have shown probabilistic model likelihoods have poor specificity for out-of-distribution (OOD) detection and often assign higher likelihoods to OOD data than in-distribution data. To ameliorate this issue we propose DoSE, the density of states estimator. Drawing on the statistical physics notion of ``density of states,'' the DoSE decision rule avoids direct comparison of model probabilities, and instead utilizes the ``probability of the model probability,'' or indeed the frequency of any reasonable statistic. The frequency is calculated using nonparametric density estimators (e.g., KDE and one-class SVM) which measure the typicality of various model statistics given the training data and from which we can flag test points with low typicality as anomalous. Unlike many other methods, DoSE requires neither labeled data nor OOD examples. DoSE is modular and can be trivially applied to any existing, trained model. We demonstrate DoSE's state-of-the-art performance against other unsupervised OOD detectors on previously established ``hard'' benchmarks.

Key facts. Cited by 74 papers in the map (data through 7 Oct 2026). First posted on arXiv 16 Jun 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 20; 2022: 13; 2023: 7; 2024: 12; 2025: 7; 2026: 15.

74 papers.

Sort:

By area:

  1. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 529
  2. 2021A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detectioncited by 206
  3. 2021DEUP: Direct Epistemic Uncertainty Predictioncited by 102
  4. 2022Denoising diffusion models for out-of-distribution detectioncited by 102
  5. 2021Understanding Failures in Out-of-Distribution Detection with Deep Generative Modelscited by 89
  6. 2021Graph Posterior Network: Bayesian Predictive Uncertainty for Node Classificationcited by 79
  7. 2021Does Your Dermatology Classifier Know What It Doesn't Know? Detecting the Long-Tail of Unseen Conditionscited by 68
  8. 2023Unsupervised Out-of-Distribution Detection with Diffusion Inpaintingcited by 57
  9. 2021Neural Mean Discrepancy for Efficient Out-of-Distribution Detectioncited by 42
  10. 2024A Geometric Explanation of the Likelihood OOD Detection Paradoxcited by 22
  11. 2024Out-of-Distribution Detection with a Single Unconditional Diffusion Modelcited by 22
  12. 2021Reliable and Trustworthy Machine Learning for Health Using Dataset Shift Detectioncited by 21
38 more
  1. 2023Sensitivity-Aware Amortized Bayesian Inferencecited by 20
  2. 2022A Learning Based Hypothesis Test for Harmful Covariate Shiftcited by 18
  3. 2021On Out-of-distribution Detection with Energy-based Modelscited by 17
  4. 2021Bayesian OOD detection with aleatoric uncertainty and outlier exposurecited by 16
  5. 2021Task-agnostic Continual Learning with Hybrid Probabilistic Modelscited by 16
  6. 2021Entropic Issues in Likelihood-Based OOD Detectioncited by 16
  7. 2022Adversarial vulnerability of powerful near out-of-distribution detectioncited by 16
  8. 2022Model-agnostic out-of-distribution detection using combined statistical testscited by 14
  9. 2025Out-of-Distribution Detection Methods Answer the Wrong Questionscited by 14
  10. 2021Natural Posterior Network: Deep Bayesian Uncertainty for Exponential Family Distributionscited by 13
  11. 2022Out-of-Distribution Detection with Reconstruction Error and Typicality-based Penaltycited by 11
  12. 2023Likelihood-based Out-of-Distribution Detection with Denoising Diffusion Probabilistic Modelscited by 10
  13. 2024Forte : Finding Outliers with Representation Typicality Estimationcited by 9
  14. 2022Boosting Out-of-Distribution Detection with Multiple Pre-trained Modelscited by 7
  15. 2021Detecting Anomalous Event Sequences with Temporal Point Processescited by 6
  16. 2022Detect, Distill and Update: Learned DB Systems Facing Out of Distribution Datacited by 6
  17. 2024CLIP-driven Outliers Synthesis for few-shot OOD detectioncited by 6
  18. 2025K4: Online Log Anomaly Detection Via Unsupervised Typicality Learningcited by 6
  19. 2021Test for non-negligible adverse shiftscited by 5
  20. 2021Data Invariants to Understand Unsupervised Out-of-Distribution Detectioncited by 5
  21. 2022A Comprehensive Review of Trends, Applications and Challenges In Out-of-Distribution Detectioncited by 5
  22. 2021Towards Robust Active Feature Acquisitioncited by 4
  23. 2023Morse Neural Networks for Uncertainty Quantificationcited by 4
  24. 2025EigenScore: OOD Detection using Covariance in Diffusion Modelscited by 4
  25. 2021InFlow: Robust outlier detection utilizing Normalizing Flowscited by 3
  26. 2024Distribution Guided Active Feature Acquisitioncited by 3
  27. 2025SCOPED: Score-Curvature Out-of-distribution Proximity Evaluator for Diffusioncited by 3
  28. 2026Trust The Typicalcited by 3
  29. 2022Meta-learning for Out-of-Distribution Detection via Density Estimation in Latent Spacecited by 2
  30. 2024Approximations to the Fisher Information Metric of Deep Generative Models for Out-Of-Distribution Detectioncited by 2
  31. 2025Zero-Shot Image Anomaly Detection Using Generative Foundation Modelscited by 2
  32. 2026Geometry over Density: Few-Shot Cross-Domain OOD Detectioncited by 2
  33. 2022A Field of Experts Prior for Adapting Neural Networks at Test Timecited by 1
  34. 2022Out-of-Distribution Detection using BiGAN and MDLcited by 1
  35. 2022Know Your Space: Inlier and Outlier Construction for Calibrating Medical OOD Detectorscited by 1
  36. 2023Master's Thesis: Out-of-distribution Detection with Energy-based Modelscited by 1
  37. 2023Building One-class Detector for Anything: Open-vocabulary Zero-shot OOD Detection Using Text-image Modelscited by 1
  38. 2023SR-OOD: Out-of-Distribution Detection via Sample Repairingcited by 1

50 most cited of 74 All 74 (JSON)