Is Task-Specific Training Necessary for Anomaly Detection?
Not on the map. Why?
Abstract (arXiv)
Current state-of-the-art multi-class unsupervised anomaly detection (MUAD) methods rely on training encoder--decoder models to reconstruct anomaly-free features. However, we argue that such task-specific training is costly under distribution shifts, and that reconstruction-based residual scoring further faces a fidelity--stability dilemma. Existing training-free alternatives, in turn, remain prone to cross-category and cross-region mismatches in MUAD. Motivated by these limitations, we propose Retrieval-based Anomaly Detection (RAD), a task-specific training-free framework that stores anomaly-free features in a memory and detects anomalies through multi-level retrieval, matching test patches against the memory. Experiments demonstrate that RAD achieves state-of-the-art performance across four established benchmarks (MVTec-AD, VisA, Real-IAD, 3D-ADAM) under both standard and few-shot settings. On MVTec-AD, RAD reaches 96.7% Pixel AUROC with just a single anomaly-free image compared to 98.5% of RAD's full-data performance. Collectively, these findings overturn the assumption that MUAD requires task-specific training, showing that state-of-the-art anomaly detection is feasible with training-free memory-based retrieval. Our code is available at https://github.com/longkukuhi/RAD.
Key facts. Cited by 2 papers in the map (data through 13 Oct 2026). Builds on 46 of its 59 references that are in the map. First posted on arXiv 30 Jan 2026.
Builds on
46 of 59 references.
References · 46
- 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,857
- 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,364
- 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,915
- 2020Training data-efficient image transformers & distillation through attentioncited by 4,574
- 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,368
- 2021An Empirical Study of Training Self-Supervised Vision Transformerscited by 1,612
- 2021iBOT: Image BERT Pre-Training with Online Tokenizercited by 895
- 2021Towards Total Recall in Industrial Anomaly Detectioncited by 774
- 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 593
- 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 530
- 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 511
- 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 420
34 more
- 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
- 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 399
- 2019Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddingscited by 386
- 2023WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentationcited by 314
- 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 299
- 2022BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizerscited by 298
- 2020Multiresolution Knowledge Distillation for Anomaly Detectioncited by 285
- 2022A Unified Model for Multi-class Anomaly Detectioncited by 268
- 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
- 2022DeSTSeg: Segmentation Guided Denoising Student-Teacher for Anomaly Detectioncited by 153
- 2024Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detectioncited by 139
- 2019Towards Visually Explaining Variational Autoencoderscited by 113
- 2024MambaAD: Exploring State Space Models for Multi-class Unsupervised Anomaly Detectioncited by 113
- 2020Deep Nearest Neighbor Anomaly Detectioncited by 104
- 2024PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detectioncited by 98
- 2024Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detectioncited by 93
- 2023What Do Self-Supervised Vision Transformers Learn?cited by 91
- 2021Divide-and-Assemble: Learning Block-wise Memory for Unsupervised Anomaly Detectioncited by 83
- 2022PNI : Industrial Anomaly Detection using Position and Neighborhood Informationcited by 69
- 2023Hierarchical Vector Quantized Transformer for Multi-class Unsupervised Anomaly Detectioncited by 66
- 2023Anomaly Detection with Conditioned Denoising Diffusion Modelscited by 56
- 2020DFR: Deep Feature Reconstruction for Unsupervised Anomaly Segmentationcited by 54
- 2021Can stable and accurate neural networks be computed? -- On the barriers of deep learning and Smale's 18th problemcited by 53
- 2023Exploring Plain ViT Reconstruction for Multi-class Unsupervised Anomaly Detectioncited by 47
- 2023ReConPatch : Contrastive Patch Representation Learning for Industrial Anomaly Detectioncited by 30
- 2024Look Inside for More: Internal Spatial Modality Perception for 3D Anomaly Detectioncited by 30
- 2024GeneralAD: Anomaly Detection Across Domains by Attending to Distorted Featurescited by 29
- 2025The MVTec AD 2 Dataset: Advanced Scenarios for Unsupervised Anomaly Detectioncited by 29
- 2024UniVAD: A Training-free Unified Model for Few-shot Visual Anomaly Detectioncited by 28
- 2024ResAD: A Simple Framework for Class Generalizable Anomaly Detectioncited by 25
- 2023Rejuvenating image-GPT as Strong Visual Representation Learnerscited by 16
- 2024Continuous Memory Representation for Anomaly Detectioncited by 16
- 2025A Lightweight 3D Anomaly Detection Method with Rotationally Invariant Featurescited by 5
- 20253D-ADAM: A Dataset for 3D Anomaly Detection in Additive Manufacturingcited by 1
Built on by
Citing papers per year: 2026: 2.
2 papers.