Generalized Out-of-Distribution Detection: A Survey
2021/10/21 by Jingkang Yang, Yang, Jingkang, Kaiyang Zhou +5 · 1 voice · 171 citations
Computer Science · Medicine · #Adversarial Robustness in Machine Learning #Anomaly Detection Techniques and Applications #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #Data-Driven Disease Surveillance #FOS: Computer and information sciences #Machine Learning (cs.LG) #cs.AI #cs.CV #cs.LG
paper · pdf · doi:10.48550/arxiv.2110.11334
openalex publication_date 2021/10/21 · arxiv published 2021/10/21 · arxiv updated 2024/01/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Abstract
Out-of-distribution (OOD) detection is critical to ensuring the reliability and safety of machine learning systems. For instance, in autonomous driving, we would like the driving system to issue an alert and hand over the control to humans when it detects unusual scenes or objects that it has never seen during training time and cannot make a safe decision. The term, OOD detection, first emerged in 2017 and since then has received increasing attention from the research community, leading to a plethora of methods developed, ranging from classification-based to density-based to distance-based ones. Meanwhile, several other problems, including anomaly detection (AD), novelty detection (ND), open set recognition (OSR), and outlier detection (OD), are closely related to OOD detection in terms of motivation and methodology. Despite common goals, these topics develop in isolation, and their subtle differences in definition and problem setting often confuse readers and practitioners. In this survey, we first present a unified framework called generalized OOD detection, which encompasses the five aforementioned problems, i.e., AD, ND, OSR, OOD detection, and OD. Under our framework, these five problems can be seen as special cases or sub-tasks, and are easier to distinguish. We then review each of these five areas by summarizing their recent technical developments, with a special focus on OOD detection methodologies. We conclude this survey with open challenges and potential research directions.
Citations
Cited by
- Invariant Feature Extraction Through Conditional Independence and the Optimal Transport Barycenter Problem: the Gaussian case
- Out-of-Distribution Detection in Molecular Complexes via Diffusion Models for Irregular Graphs
- Few-Shot Learning of a Graph-Based Neural Network Model Without Backpropagation
- MAD-OOD: A Deep Learning Cluster-Driven Framework for an Out-of-Distribution Malware Detection and Classification
- Open Ad-hoc Categorization with Contextualized Feature Learning
- Out-of-Distribution Detection for Continual Learning: Design Principles and Benchmarking
- Improving the Accuracy of Amortized Model Comparison with Self-Consistency
- General OOD Detection via Model-aware and Subspace-aware Variable Priority
- Rethinking Jailbreak Detection of Large Vision Language Models with Representational Contrastive Scoring
- On the Dangers of Bootstrapping Generation for Continual Learning and Beyond
- Forests of Uncertaint(r)ees: Using tree-based ensembles to estimate probability distributions of future conflict
- Knowing when to trust machine-learned interatomic potentials
- How (Mis)calibrated is Your Federated CLIP and What To Do About It?
- Domain Feature Collapse: Implications for Out-of-Distribution Detection and Solutions
- Continual Error Correction on Low-Resource Devices
- Uncertainty of Network Topology with Applications to Out-of-Distribution Detection
- Fast Post-Hoc Confidence Fusion for 3-Class Open-Set Aerial Object Detection
- Latent space analysis and generalization to out-of-distribution data
- Ellipsoid-Based Decision Boundaries for Open Intent Classification
- FedSDWC: Federated Synergistic Dual-Representation Weak Causal Learning for OOD
- Automatic Grid Updates for Kolmogorov-Arnold Networks using Layer Histograms
- Relative Energy Learning for LiDAR Out-of-Distribution Detection
- Jr. AI Scientist and Its Risk Report: Autonomous Scientific Exploration from a Baseline Paper
- Covariance Last-Layer Ensembles: Function-Space Diversity for Efficient Uncertainty Quantification
- Improving Out-of-Distribution Detection via Dynamic Covariance Calibration
- Face-Trace: Open-Set Attribution and Progressive Discovery of Synthetic Face Generators
- Jaeger: an accurate and fast deep-learning tool to detect bacteriophage sequences
- Instance-Aware Pseudo-Labeling and Class-Focused Contrastive Learning for Weakly Supervised Domain Adaptive Segmentation of Electron Microscopy
- Anomaly detection for high-content image-based phenotypic cell profiling
- Humanoid-inspired Causal Representation Learning for Domain Generalization
- In Search of the Unknown Unknowns: A Multi-Metric Distance Ensemble for Out of Distribution Anomaly Detection in Astronomical Surveys
- Floating-Point Neural Network Verification at the Software Level
- Worst-Case Distance-Aware Error Bounds for Neural Networks
- Out-of-Distribution Detection for Safety Assurance of AI and Autonomous Systems
- Exploring Scale Shift in Crowd Localization under the Context of Domain Generalization
- Ensuring Robustness in ML-enabled Software Systems: A User Survey
- Can VLMs Unlock Semantic Anomaly Detection? A Framework for Structured Reasoning
- Nearest-Class Mean and Logits Agreement for Wildlife Open-Set Recognition
- Learning After Model Deployment
- GOOD: Training-Free Guided Diffusion Sampling for Out-of-Distribution Detection
- Benchmarking Out-of-Distribution Detection for Plankton Recognition: A Systematic Evaluation of Advanced Methods in Marine Ecological Monitoring
- A Geometry-Based View of Mahalanobis OOD Detection
- A Multi-dimensional Semantic Surprise Framework Based on Low-Entropy Semantic Manifolds for Fine-Grained Out-of-Distribution Detection
- Hierarchical Reasoning with Vision-Language Models for Incident Reports from Dashcam Videos
- Artificial intelligence as a surrogate brain: Bridging neural dynamical models and data
- A Generic Machine Learning Framework for Radio Frequency Fingerprinting
- Failure Prediction at Runtime for Generative Robot Policies
- EigenScore: OOD Detection using Covariance in Diffusion Models
- Medix: Out-of-Distribution Detection from Unlabeled Wild Data via Robust Gradient Statistics
- Out-of-Distribution Detection from Small Training Sets using Bayesian Neural Network Classifiers
- Can multi-label classification networks know what they don't know?
- Generalized Fine-Grained Category Discovery with Multi-Granularity Conceptual Experts
- Rotation Control Unlearning: Quantifying and Controlling Continuous Unlearning for LLM with The Cognitive Rotation Space
- DynaSubVAE: Adaptive Subgrouping for Scalable and Robust OOD Detection
- Exploring Transition States of Protein Conformational Changes via Out-of-Distribution Detection in the Hyperspherical Latent Space
- LABELING COPILOT: A Deep Research Agent for Automated Data Curation in Computer Vision
- Category Discovery: An Open-World Perspective
- Multidimensional Uncertainty Quantification via Optimal Transport
- Less Precise Can Be More Reliable: A Systematic Evaluation of Quantization's Impact on VLMs Beyond Accuracy
- The Use of the Simplex Architecture to Enhance Safety in Deep-Learning-Powered Autonomous Systems
- ICONIC-444: A 3.1-Million-Image Dataset for OOD Detection Research
- Benchmarking Foundation and Large Language Models for Few-Shot Medical Image Segmentation
- PGTuner: An Efficient Framework for Automatic and Transferable Configuration Tuning of Proximity Graphs
- Linear Regression under Missing or Corrupted Coordinates
- Probabilistic Runtime Verification, Evaluation and Risk Assessment of Visual Deep Learning Systems
- Text Meets Topology: Rethinking Out-of-distribution Detection in Text-Rich Networks
- Long-Tailed Out-of-Distribution Detection with Refined Separate Class Learning
- An Empirical Analysis of VLM-based OOD Detection: Mechanisms, Advantages, and Sensitivity
- Out of Distribution Detection in Self-adaptive Robots with AI-powered Digital Twins
- Sharpness-Aware Geometric Defense for Robust Out-Of-Distribution Detection
- Prompt Optimization Meets Subspace Representation Learning for Few-shot Out-of-Distribution Detection
- Tackling the Noisy Elephant in the Room: Label Noise-robust Out-of-Distribution Detection via Loss Correction and Low-rank Decomposition
- DCV-ROOD Evaluation Framework: Dual Cross-Validation for Robust Out-of-Distribution Detection
- Prior Distribution and Model Confidence
- Towards Open World Detection: A Survey
- Improving the Resilience of Quadrotors in Underground Environments by Combining Learning-based and Safety Controllers
- An Investigation of Visual Foundation Models Robustness
- Frustratingly Easy Feature Reconstruction for Out-of-Distribution Detection
- Domain Adaptation via Feature Refinement
- Novel Category Discovery with X-Agent Attention for Open-Vocabulary Semantic Segmentation
- Energy Landscapes Enable Reliable Abstention in Retrieval-Augmented Large Language Models for Healthcare
- What Can We Learn from Harry Potter? An Exploratory Study of Visual Representation Learning from Atypical Videos
- Activation Subspaces for Out-of-Distribution Detection
- Probability Density from Latent Diffusion Models for Out-of-Distribution Detection
- Improving the Accuracy of Amortized Model Comparison with Self-Consistency
- Empirical Evidences for the Effects of Feature Diversity in Open Set Recognition and Continual Learning
- M3OOD: Automatic Selection of Multimodal OOD Detectors
- Out-of-Distribution Detection using Counterfactual Distance
- How Safe Will I Be Given What I Saw? Calibrated Prediction of Safety Chances for Image-Controlled Autonomy
- Hierarchical Adaptive networks with Task vectors for Test-Time Adaptation
- Deep learning methods for 2D material electronic properties
- Detecting Model Misspecification in Cosmology with Scale-Dependent Normalizing Flows
- Pseudo-label Induced Subspace Representation Learning for Robust Out-of-Distribution Detection
- Infrared Object Detection with Ultra Small ConvNets: Is ImageNet Pretraining Still Useful?
- Superior resilience to poisoning and amenability to unlearning in quantum machine learning
- BOOD: Boundary-based Out-Of-Distribution Data Generation
- Out-of-Distribution Detection in Medical Imaging via Diffusion Trajectories
- Challenges of Trustworthy Federated Learning: What's Done, Current Trends and Remaining Work
- Zero-Shot Image Anomaly Detection Using Generative Foundation Models
- COOkeD: Ensemble-based OOD detection in the era of zero-shot CLIP
- Foundation Models and Transformers for Anomaly Detection: A Survey
- Handling Out-of-Distribution Data: A Survey
- Direct-to-Consumer AI Health Services: Precision Healthcare Needs Precision Consent
- Hallucination Detection and Mitigation with Diffusion in Multi-Variate Time-Series Foundation Models
- Integration of Old and New Knowledge for Generalized Intent Discovery: A Consistency-driven Prototype-Prompting Framework
- NegRefine: Refining Negative Label-Based Zero-Shot OOD Detection
- PatchGuard: Adversarially Robust Anomaly Detection and Localization through Vision Transformers and Pseudo Anomalies
- ADAPT: A Pseudo-labeling Approach to Combat Concept Drift in Malware Detection
- Feature Bank Enhancement for Distance-based Out-of-Distribution Detection
- Last Layer Hamiltonian Monte Carlo
- DIsoN: Decentralized Isolation Networks for Out-of-Distribution Detection in Medical Imaging
- ViLU: Learning Vision-Language Uncertainties for Failure Prediction
- AI Should Sense Better, Not Just Scale Bigger: Adaptive Sensing as a Paradigm Shift
- Physics-Grounded Motion Forecasting via Equation Discovery for Trajectory-Guided Image-to-Video Generation
- Understanding Knowledge Transferability for Transfer Learning: A Survey
- Large Language Models for Crash Detection in Video: A Survey of Methods, Datasets, and Challenges
- Gradient Short-Circuit: Efficient Out-of-Distribution Detection via Feature Intervention
- Out-of-Distribution Detection Methods Answer the Wrong Questions
- Spurious-Aware Prototype Refinement for Reliable Out-of-Distribution Detection
- The Illusion of Progress? A Critical Look at Test-Time Adaptation for Vision-Language Models
- Detecting What Matters: A Novel Approach for Out-of-Distribution 3D Object Detection in Autonomous Vehicles
- Adaptive Multi-prompt Contrastive Network for Few-shot Out-of-distribution Detection
- Keeping Medical AI Healthy and Trustworthy: A Review of Detection and Correction Methods for System Degradation
- LAION-C: An Out-of-Distribution Benchmark for Web-Scale Vision Models
- Energy-Based Transfer for Reinforcement Learning
- FOCoOp: Enhancing Out-of-Distribution Robustness in Federated Prompt Learning for Vision-Language Models
- TRUST: Transparent, Robust and Ultra-Sparse Trees
- Enclosing Prototypical Variational Autoencoder for Explainable Out-of-Distribution Detection
- FindMeIfYouCan: Bringing Open Set metrics to near , far and farther Out-of-Distribution Object Detection
- Improving Wildlife Out-of-Distribution Detection: Africas Big Five
- No Data? No Problem: Synthesizing Security Graphs for Better Intrusion Detection
- Neurosymbolic Artificial Intelligence for Robust Network Intrusion Detection: From Scratch to Transfer Learning
- Enhancing Abnormality Identification: Robust Out-of-Distribution Strategies for Deepfake Detection
- KairosAD: A SAM-Based Model for Industrial Anomaly Detection on Embedded Devices
- Invariant Link Selector for Spatial-Temporal Out-of-Distribution Problem
- Navigating the Latent Space Dynamics of Neural Models
- Improving Out-of-Distribution Detection with Markov Logic Networks
- Familiarity-Based Open-Set Recognition Under Adversarial Attacks
- When Shift Happens - Confounding Is to Blame
- ASAT: Adaptive Scoring and Thresholding with Human Feedback for Robust Out-of-Distribution Detection
- Recent Advances in Out-of-Distribution Detection with CLIP-Like Models: A Survey
- What Is AI Safety? What Do We Want It to Be?
- Robustness in Large Language Models: A Survey of Mitigation Strategies and Evaluation Metrics
- TULiP: Test-time Uncertainty Estimation via Linearization and Weight Perturbation
- Freeze, Prompt, and Adapt: A Framework for Source-free Unsupervised GNN Prompting
- GradPCA: Leveraging NTK Alignment for Reliable Out-of-Distribution Detection
- Refining Neural Activation Patterns for Layer-Level Concept Discovery in Neural Network-Based Receivers
- Kernel PCA for Out-of-Distribution Detection: Non-Linear Kernel Selections and Approximations
- SpectralGap: Graph-Level Out-of-Distribution Detection via Laplacian Eigenvalue Gaps
- LoD: Loss-difference OOD Detection by Intentionally Label-Noisifying Unlabeled Wild Data
- GEM: Gaussian Embedding Modeling for Out-of-Distribution Detection in GUI Agents
- Training-Free Out-of-Distribution Detection for Pathology Whole-Slide Images
- Unsupervised Detection of Distribution Shift in Inverse Problems using Diffusion Models
- Open Set Domain Adaptation with Vision-language models via Gradient-aware Separation
- Is Supervised Learning Really That Different from Unsupervised?
- Does Feasibility Matter? Understanding the Impact of Feasibility on Synthetic Training Data
- Uncertainty-aware Latent Safety Filters for Avoiding Out-of-Distribution Failures
- A Deployment Audit of Release-Side Risk in Conformal Triage under Prevalence Shift
- SimPRIVE: a Simulation framework for Physical Robot Interaction with Virtual Environments
- Graph Synthetic Out-of-Distribution Exposure with Large Language Models
- Out-of-Distribution Detection in Heterogeneous Graphs via Energy Propagation
- GLIP-OOD: Zero-Shot Graph OOD Detection with Graph Foundation Model
- Digital Twin-based Out-of-Distribution Detection in Autonomous Vessels
- Segmenting Objectiveness and Task-awareness Unknown Region for Autonomous Driving
- ISEE: Interactive Semantic Enrichment for Database Fields
- Dream-Box: Object-wise Outlier Generation for Out-of-Distribution Detection
- Comparing Uncertainty Measurement and Mitigation Methods for Large Language Models: A Systematic Review
- Rare Event Analysis of Large Language Models
- SABRE: A Multi-Agent Approach for Selecting Out-of-Distribution Detectors Under a Budget
- Can We Ignore Labels In Out of Distribution Detection?
- Proxy-Anchor and EVT-Driven Continual Learning Method for Generalized Category Discovery
- Reducing Class Bias In Data-Balanced Datasets Through Hardness-Based Resampling
- DebGCD: Debiased Learning with Distribution Guidance for Generalized Category Discovery
Discussions
Related