mlsanityData to 14 Oct 2026

Enhancing Out-of-Distribution Detection with Extended Logit Normalization

arXiv 2504.11434 ↗First posted 15 Apr 2025cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

noindent Out-of-distribution (OOD) detection is essential for the safe deployment of machine learning models. Extensive work has focused on devising various scoring functions for detecting OOD samples, while only a few studies focus on training neural networks using certain model calibration objectives, which often lead to a compromise in predictive accuracy and support only limited choices of scoring functions. In this work, we first identify the feature collapse phenomena in Logit Normalization (LogitNorm), then propose a novel hyperparameter-free formulation that significantly benefits a wide range of post-hoc detection methods. To be specific, we devise a feature distance-awareness loss term in addition to LogitNorm, termed ELogitNorm, which enables improved OOD detection and in-distribution (ID) confidence calibration. Extensive experiments across standard benchmarks demonstrate that our approach outperforms state-of-the-art training-time methods in OOD detection while maintaining strong ID classification accuracy. Our code is available on: https://github.com/limchaos/ElogitNorm.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 51 of its 62 references that are in the map. First posted on arXiv 15 Apr 2025.

Builds on

51 of 62 references.

References · 51

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  4. 2017On Calibration of Modern Neural Networkscited by 4,960
  5. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,898
  6. 2020Supervised Contrastive Learningcited by 2,829
  7. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
  8. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,636
  9. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  10. 2017Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networkscited by 1,328
  11. 2020Energy-based Out-of-distribution Detectioncited by 1,299
  12. 2018Deep Anomaly Detection with Outlier Exposurecited by 942
38 more
  1. 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 776
  2. 2022Out-of-Distribution Detection with Deep Nearest Neighborscited by 600
  3. 2021ReAct: Out-of-distribution Detection With Rectified Activationscited by 486
  4. 2019Scaling Out-of-Distribution Detection for Real-World Settingscited by 465
  5. 2017Regularizing Neural Networks by Penalizing Confident Output Distributionscited by 464
  6. 2018Do Deep Generative Models Know What They Don't Know?cited by 443
  7. 2020CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instancescited by 441
  8. 2020Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Datacited by 408
  9. 2022ViM: Out-Of-Distribution with Virtual-logit Matchingcited by 405
  10. 2021Understanding Dimensional Collapse in Contrastive Self-supervised Learningcited by 378
  11. 2022VOS: Learning What You Don't Know by Virtual Outlier Synthesiscited by 300
  12. 2021Exploring the Limits of Out-of-Distribution Detectioncited by 283
  13. 2022Mitigating Neural Network Overconfidence with Logit Normalizationcited by 252
  14. 2022Extremely Simple Activation Shaping for Out-of-Distribution Detectioncited by 212
  15. 2020Mix-n-Match: Ensemble and Compositional Methods for Uncertainty Calibration in Deep Learningcited by 190
  16. 2020Contrastive Training for Improved Out-of-Distribution Detectioncited by 186
  17. 2023OpenOOD v1.5: Enhanced Benchmark for Out-of-Distribution Detectioncited by 179
  18. 2021On Feature Decorrelation in Self-Supervised Learningcited by 160
  19. 2022RankMe: Assessing the downstream performance of pretrained self-supervised representations by their rankcited by 153
  20. 2023Non-Parametric Outlier Synthesiscited by 129
  21. 2024Addressing Representation Collapse in Vector Quantized Models with One Linear Layercited by 80
  22. 2024Benchmarking Uncertainty Disentanglement: Specialized Uncertainties for Specialized Taskscited by 78
  23. 2023Dream the Impossible: Outlier Imagination with Diffusion Modelscited by 77
  24. 2023Scaling for Training Time and Post-hoc Out-of-distribution Detection Enhancementcited by 77
  25. 2023Nearest Neighbor Guidance for Out-of-Distribution Detectioncited by 64
  26. 2023Feed Two Birds with One Scone: Exploiting Wild Data for Both Out-of-Distribution Generalization and Detectioncited by 57
  27. 2022RankFeat: Rank-1 Feature Removal for Out-of-distribution Detectioncited by 52
  28. 2023Fast Decision Boundary based Out-of-Distribution Detectorcited by 49
  29. 2021Parameterized Temperature Scaling for Boosting the Expressive Power in Post-Hoc Uncertainty Calibrationcited by 48
  30. 2022Understanding Collapse in Non-Contrastive Siamese Representation Learningcited by 42
  31. 2023NECO: NEural Collapse Based Out-of-distribution detectioncited by 42
  32. 2022Block Selection Method for Using Feature Norm in Out-of-distribution Detectioncited by 30
  33. 2020Why is the Mahalanobis Distance Effective for Anomaly Detection?cited by 28
  34. 2020Multi-Class Uncertainty Calibration via Mutual Information Maximization-based Binningcited by 27
  35. 2023Understanding the Feature Norm for Out-of-Distribution Detectioncited by 27
  36. 2023Proximity-Informed Calibration for Deep Neural Networkscited by 18
  37. 2022Investigating Power laws in Deep Representation Learningcited by 17
  38. 2025Revisiting Likelihood-Based Out-of-Distribution Detection by Modeling Representationscited by 6

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 1.

3 papers.

  1. 2025Out-of-Distribution Detection for Continual Learning: Design Principles and Benchmarkingcited by 1
  2. 2025Bag of Coins: A Statistical Probe into Neural Confidence Structurescited by 0
  3. 2026A Dataset is Worth 1 MBcited by 0