mlsanityData to 14 Oct 2026

Revisiting Logit Distributions for Reliable Out-of-Distribution Detection

arXiv 2510.20134 ↗First posted 23 Oct 2025cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

Out-of-distribution (OOD) detection is critical for ensuring the reliability of deep learning models in open-world applications. While post-hoc methods are favored for their efficiency and ease of deployment, existing approaches often underexploit the rich information embedded in the model's logits space. In this paper, we propose LogitGap, a novel post-hoc OOD detection method that explicitly exploits the relationship between the maximum logit and the remaining logits to enhance the separability between in-distribution (ID) and OOD samples. To further improve its effectiveness, we refine LogitGap by focusing on a more compact and informative subset of the logit space. Specifically, we introduce a training-free strategy that automatically identifies the most informative logits for scoring. We provide both theoretical analysis and empirical evidence to validate the effectiveness of our approach. Extensive experiments on both vision-language and vision-only models demonstrate that LogitGap consistently achieves state-of-the-art performance across diverse OOD detection scenarios and benchmarks. Code is available at https://github.com/GIT-LJc/LogitGap.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 43 of its 61 references that are in the map. First posted on arXiv 23 Oct 2025.

Builds on

43 of 61 references.

References · 43

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
  5. 2021Learning to Prompt for Vision-Language Modelscited by 2,281
  6. 2013Describing Textures in the Wildcited by 2,125
  7. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,850
  8. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  9. 2019Natural Adversarial Examplescited by 1,408
  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. 2019Learning Robust Global Representations by Penalizing Local Predictive Powercited by 1,013
31 more
  1. 2018Deep Anomaly Detection with Outlier Exposurecited by 942
  2. 2021ImageNet-21K Pretraining for the Massescited by 613
  3. 2022Out-of-Distribution Detection with Deep Nearest Neighborscited by 600
  4. 2021ReAct: Out-of-distribution Detection With Rectified Activationscited by 486
  5. 2019Scaling Out-of-Distribution Detection for Real-World Settingscited by 465
  6. 2020Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Datacited by 408
  7. 2022ViM: Out-Of-Distribution with Virtual-logit Matchingcited by 405
  8. 2021Open-Set Recognition: a Good Closed-Set Classifier is All You Need?cited by 353
  9. 2021On the Importance of Gradients for Detecting Distributional Shifts in the Wildcited by 339
  10. 2022VOS: Learning What You Don't Know by Virtual Outlier Synthesiscited by 300
  11. 2022OpenOOD: Benchmarking Generalized Out-of-Distribution Detectioncited by 268
  12. 2022Mitigating Neural Network Overconfidence with Logit Normalizationcited by 252
  13. 2021MOS: Towards Scaling Out-of-distribution Detection for Large Semantic Spacecited by 233
  14. 2022Delving into Out-of-Distribution Detection with Vision-Language Representationscited by 216
  15. 2022Extremely Simple Activation Shaping for Out-of-Distribution Detectioncited by 212
  16. 2023In or Out? Fixing ImageNet Out-of-Distribution Detection Evaluationcited by 134
  17. 2023CLIPN for Zero-Shot OOD Detection: Teaching CLIP to Say Nocited by 130
  18. 2019Unsupervised Out-of-Distribution Detection by Maximum Classifier Discrepancycited by 112
  19. 2021Semantically Coherent Out-of-Distribution Detectioncited by 107
  20. 2023LoCoOp: Few-Shot Out-of-Distribution Detection via Prompt Learningcited by 107
  21. 2021Mixture Outlier Exposure: Towards Out-of-Distribution Detection in Fine-grained Environmentscited by 66
  22. 2024Negative Label Guided OOD Detection with Pretrained Vision-Language Modelscited by 66
  23. 2023Fast Decision Boundary based Out-of-Distribution Detectorcited by 49
  24. 2024Learning Transferable Negative Prompts for Out-of-Distribution Detectioncited by 43
  25. 2023ID-like Prompt Learning for Few-Shot Out-of-Distribution Detectioncited by 42
  26. 2023GL-MCM: Global and Local Maximum Concept Matching for Zero-Shot Out-of-Distribution Detectioncited by 36
  27. 2023Detecting Out-of-Distribution Through the Lens of Neural Collapsecited by 34
  28. 2024Envisioning Outlier Exposure by Large Language Models for Out-of-Distribution Detectioncited by 32
  29. 2023ImageNet-OOD: Deciphering Modern Out-of-Distribution Detection Algorithmscited by 22
  30. 2024Conjugated Semantic Pool Improves OOD Detection with Pre-trained Vision-Language Modelscited by 13
  31. 2024Image-based Outlier Synthesis With Training Datacited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026Component-Based Out-of-Distribution Detectioncited by 0
  2. 2026Modality-Aware Out-of-Distribution Detection for Multi-Modal Action Recognitioncited by 0
  3. 2026Uncertainty quantification for trustworthy deep learning: Methods and measurescited by 0
  4. 2026CHILD: Human-in-the-Loop OOD Detection for Safe Clinical Deploymentcited by 0