Dataset Identity, Not Novelty: The Source of an Inflated OOD Detection Gain
cs.LG, cs.CV
Submitted: 2026-10-01
Updated: 2026-10-01
Code: https://github.com/rwightman/pytorch-image-models
Terminology
Sources
- Deep Anomaly Detection with Outlier Exposure
- MahaVar: OOD Detection via Class-wise Mahalanobis Distance Variance under Neural Collapse
- Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks
- Improving Out-of-Distribution Detection by Combining Existing Post-hoc Methods
- A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detection
- A Less Biased Evaluation of Out-of-distribution Sample Detectors
- No True State-of-the-Art? OOD Detection Methods are Inconsistent across Datasets
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks