mlsanityData to 14 Oct 2026

Anchor and Adapt: Asymmetric Prompt Adaptation for Few-Shot Industrial Anomaly Detection

arXiv 2610.07016 ↗First posted 4 Oct 2026cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

In few-shot industrial anomaly detection, the few normal target images provide no direct defect supervision, making anomaly prompts difficult to learn from these samples alone. Some vision-language methods therefore use manually specified descriptions to supply explicit anomaly semantics. However, constructing these descriptions requires product-specific effort, and their effectiveness depends on prompt selection. We propose Anchor and Adapt, a two-stage prompt learning framework that separates the acquisition of anomaly semantics from adaptation to target normal appearance. Stage I learns transferable normal and abnormal anchors from annotated auxiliary data. Stage II keeps these anchors fixed and adapts an additional normal branch using the few target normal samples. The inherited and adapted normal branches jointly characterize target normality, with text-anchor regularization encouraging consistency with the generic normal prior and separation from the abnormal anchors. This design retains learned anomaly knowledge while reducing dependence on category-specific anomaly templates, without requiring synthetic anomaly generation. Cross-dataset experiments between MVTec-AD and VisA under 1-, 2-, and 4-shot settings demonstrate competitive detection and localization performance. Controlled ablations assess the roles of transferred anchors, asymmetric adaptation, dual-normal representations, and anchor regularization.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 34 of its 49 references that are in the map. First posted on arXiv 4 Oct 2026.

Builds on

34 of 49 references.

References · 34

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,864
  3. 2021Learning to Prompt for Vision-Language Modelscited by 2,281
  4. 2022Conditional Prompt Learning for Vision-Language Modelscited by 1,433
  5. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 776
  6. 2021VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understandingcited by 602
  7. 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 594
  8. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 531
  9. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
  10. 2023WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentationcited by 315
  11. 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 300
  12. 2023AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detectioncited by 271
22 more
  1. 2019Dice Loss for Data-imbalanced NLP Taskscited by 182
  2. 2023AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Modelscited by 171
  3. 2022SegCLIP: Patch Aggregation with Learnable Centers for Open-Vocabulary Semantic Segmentationcited by 164
  4. 2024PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detectioncited by 98
  5. 2024Toward Generalist Anomaly Detection via In-context Residual Learning with Few-shot Sample Promptscited by 89
  6. 2025AA-CLIP: Enhancing Zero-shot Anomaly Detection via Anomaly-Aware CLIPcited by 73
  7. 2024VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentationcited by 50
  8. 2023Chinese Text Recognition with A Pre-Trained CLIP-Like Model Through Image-IDS Aligningcited by 26
  9. 2024ResAD: A Simple Framework for Class Generalizable Anomaly Detectioncited by 25
  10. 2024Progressive Boundary Guided Anomaly Synthesis for Industrial Anomaly Detectioncited by 17
  11. 2023Random Word Data Augmentation with CLIP for Zero-Shot Anomaly Detectioncited by 16
  12. 2025ChatReID: Open-ended Interactive Person Retrieval via Hierarchical Progressive Tuning for Vision Language Modelscited by 16
  13. 2025MetaUAS: Universal Anomaly Segmentation with One-Prompt Meta-Learningcited by 13
  14. 2025UMIT: Unifying Medical Imaging Tasks via Vision-Language Modelscited by 12
  15. 2025DictAS: A Framework for Class-Generalizable Few-Shot Anomaly Segmentation via Dictionary Lookupcited by 10
  16. 2025FiLo++: Zero-/Few-Shot Anomaly Detection by Fused Fine-Grained Descriptions and Deformable Localizationcited by 8
  17. 2025FastRef:Fast Prototype Refinement for Few-Shot Industrial Anomaly Detectioncited by 7
  18. 2024Bidirectional Multimodal Prompt Learning with Scale-Aware Training for Few-Shot Multi-Class Anomaly Detectioncited by 4
  19. 2025Towards Fine-Grained Vision-Language Alignment for Few-Shot Anomaly Detectioncited by 4
  20. 2025IADGPT: Unified LVLM for Few-Shot Industrial Anomaly Detection, Localization, and Reasoning via In-Context Learningcited by 2
  21. 2025OmniPT: Unleashing the Potential of Large Vision Language Models for Pedestrian Tracking and Understandingcited by 1
  22. 2026VD-DeepStack: Bridging Visual Comparison and Language Reasoning for Few-Shot Anomaly Detectioncited by 1

Built on by

Not cited yet.