mlsanityData to 14 Oct 2026

Detecting Out-of-Distribution Inputs to Deep Generative Models Using Typicality

arXiv 1906.02994 ↗First posted 7 Jun 2019stat.ML cs.LGcited by 31

Not on the map. Why?

Abstract (arXiv)

Recent work has shown that deep generative models can assign higher likelihood to out-of-distribution data sets than to their training data (Nalisnick et al., 2019; Choi et al., 2019). We posit that this phenomenon is caused by a mismatch between the model's typical set and its areas of high probability density. In-distribution inputs should reside in the former but not necessarily in the latter, as previous work has presumed. To determine whether or not inputs reside in the typical set, we propose a statistically principled, easy-to-implement test using the empirical distribution of model likelihoods. The test is model agnostic and widely applicable, only requiring that the likelihood can be computed or closely approximated. We report experiments showing that our procedure can successfully detect the out-of-distribution sets in several of the challenging cases reported by Nalisnick et al. (2019).

Key facts. Cited by 31 papers in the map (data through 14 Oct 2026). First posted on arXiv 7 Jun 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 10; 2022: 13; 2023: 3; 2024: 3; 2025: 1; 2026: 1.

31 papers.

Sort:

By area:

  1. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 531
  2. 2023Unsupervised Out-of-Distribution Detection with Diffusion Inpaintingcited by 59
  3. 2021Score-Based Generative Classifierscited by 53
  4. 2021On the Out-of-distribution Generalization of Probabilistic Image Modellingcited by 33
  5. 2021Conditional Invertible Neural Networks for Diverse Image-to-Image Translationcited by 24
  6. 2023Distribution Shift Inversion for Out-of-Distribution Predictioncited by 23
  7. 2021Reliable and Trustworthy Machine Learning for Health Using Dataset Shift Detectioncited by 21
  8. 2021Do We Really Need to Learn Representations from In-domain Data for Outlier Detection?cited by 16
  9. 2022Model-agnostic out-of-distribution detection using combined statistical testscited by 14
  10. 2022Variational Inference MPC using Normalizing Flows and Out-of-Distribution Projectioncited by 13
  11. 2022CADet: Fully Self-Supervised Out-Of-Distribution Detection With Contrastive Learningcited by 11
  12. 2022Task Agnostic and Post-hoc Unseen Distribution Detectioncited by 8
19 more
  1. 2022Falsehoods that ML researchers believe about OOD detectioncited by 8
  2. 2023ODIM: Outlier Detection via Likelihood of Under-Fitted Generative Modelscited by 7
  3. 2021Detecting Anomalous Event Sequences with Temporal Point Processescited by 6
  4. 2021Data Invariants to Understand Unsupervised Out-of-Distribution Detectioncited by 5
  5. 2022Exploiting Inter-Sample Affinity for Knowability-Aware Universal Domain Adaptationcited by 5
  6. 2021Multi-Resolution Continuous Normalizing Flowscited by 4
  7. 2022Out-of-Distribution Detection with Class Ratio Estimationcited by 4
  8. 2024Nystr"om Kernel Stein Discrepancycited by 4
  9. 2022Self-Supervised Anomaly Detection by Self-Distillation and Negative Samplingcited by 3
  10. 2021Generative Max-Mahalanobis Classifiers for Image Classification, Generation and Morecited by 1
  11. 2022Learning by Erasing: Conditional Entropy based Transferable Out-Of-Distribution Detectioncited by 1
  12. 2022Shaken, and Stirred: Long-Range Dependencies Enable Robust Outlier Detection with PixelCNN++cited by 1
  13. 2025Sparse, self-organizing ensembles of local kernels detect rare statistical anomaliescited by 1
  14. 2022Provably Uncertainty-Guided Universal Domain Adaptationcited by 0
  15. 2022Pseudo-OOD training for robust language modelscited by 0
  16. 2022On the Connection of Generative Models and Discriminative Models for Anomaly Detectioncited by 0
  17. 2024Learning Non-Linear Invariants for Unsupervised Out-of-Distribution Detectioncited by 0
  18. 2024Resultant: Incremental Effectiveness on Likelihood for Unsupervised Out-of-Distribution Detectioncited by 0
  19. 2026When Less is More: Understanding When Token Filtering Helps and Fails in AI-generated Text Detectioncited by 0