mlsanityData to 14 Oct 2026

DeCoFlow: Structural Decomposition of Normalizing Flows for Continual Anomaly Detection

arXiv 2606.26687 ↗First posted 25 Jun 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

In industrial environments, new product categories arrive sequentially, requiring continual anomaly detection without access to past data. Normalizing Flows (NFs) provide exact density estimation but suffer from catastrophic forgetting as parameter updates across tasks distort the density manifold. While parameter isolation can prevent interference, it must preserve the strict invertibility and Jacobian validity of NFs. To satisfy these requirements, we exploit the inherent property that affine coupling layers maintain transformation validity regardless of subnet parameterization. Based on this, we propose DeCoFlow, which decomposes subnets into a frozen universal base and task-specific low-rank adapters to isolate updates. We further introduce Task-Specific Alignment, Auxiliary Coupling Layers, and Tail-Aware Loss to compensate for frozen-base rigidity. DeCoFlow achieves state-of-the-art image-level AUROCs of 98.40% on MVTec-AD and 93.00% on VisA, while maintaining parameter-level zero forgetting (0.00% FM under correct routing) with only 2.27M parameters per task.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 33 of its 42 references that are in the map. First posted on arXiv 25 Jun 2026.

Builds on

33 of 42 references.

References · 33

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  3. 2016Learning without Forgettingcited by 2,496
  4. 2016Density estimation using Real NVPcited by 2,042
  5. 2017Gradient Episodic Memory for Continual Learningcited by 1,898
  6. 2016Progressive Neural Networkscited by 1,361
  7. 2017Continual Learning with Deep Generative Replaycited by 1,158
  8. 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,105
  9. 2017PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruningcited by 898
  10. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 776
  11. 2018Overcoming catastrophic forgetting with hard attention to the taskcited by 681
  12. 2016Training Region-based Object Detectors with Online Hard Example Miningcited by 579
21 more
  1. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 421
  2. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
  3. 2021CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flowscited by 321
  4. 2022A Unified Model for Multi-class Anomaly Detectioncited by 268
  5. 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
  6. 2021FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flowscited by 227
  7. 2022CFA: Coupled-hypersphere-based Feature Adaptation for Target-Oriented Anomaly Localizationcited by 135
  8. 2024Unsupervised Continual Anomaly Detection with Contrastively-learned Promptcited by 31
  9. 2024Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detectioncited by 28
  10. 2023MSFlow: Multi-Scale Flow-based Framework for Unsupervised Anomaly Detectioncited by 25
  11. 2023An Incremental Unified Framework for Small Defect Inspectioncited by 25
  12. 2025One-for-More: Continual Diffusion Model for Anomaly Detectioncited by 19
  13. 2025MINGLE: Mixture of Null-Space Gated Low-Rank Experts for Test-Time Continual Model Mergingcited by 14
  14. 2025Gated Integration of Low-Rank Adaptation for Continual Learning of Large Language Modelscited by 13
  15. 2025ReplayCAD: Generative Diffusion Replay for Continual Anomaly Detectioncited by 9
  16. 2025Continuous Subspace Optimization for Continual Learningcited by 9
  17. 2025AnaCP: Toward Upper-Bound Continual Learning via Analytic Contrastive Projectioncited by 7
  18. 2024VQ-Flow: Taming Normalizing Flows for Multi-Class Anomaly Detection via Hierarchical Vector Quantizationcited by 5
  19. 2023Look At Me, No Replay! SurpriseNet: Anomaly Detection Inspired Class Incremental Learningcited by 2
  20. 2025Multimodal Task Representation Memory Bank vs. Catastrophic Forgetting in Anomaly Detectioncited by 2
  21. 2025CADIC: Continual Anomaly Detection Based on Incremental Coresetcited by 2

Built on by

Not cited yet.