mlsanityData to 14 Oct 2026

SOWA: Adapting Hierarchical Frozen Window Self-Attention to Visual-Language Models for Better Anomaly Detection

arXiv 2407.03634 ↗First posted 4 Jul 2024cs.CVcited by 2

On the map: Multimodal LLMs / vision-language models → Chart understanding and anomaly detection

Abstract (arXiv)

Visual anomaly detection is essential in industrial manufacturing, yet traditional methods often rely heavily on extensive normal datasets and task-specific models, limiting their scalability. Recent advancements in large-scale vision-language models have significantly enhanced zero- and few-shot anomaly detection. However, these approaches may not fully leverage hierarchical features, potentially overlooking nuanced details crucial for accurate detection. To address this, we introduce a novel window self-attention mechanism based on the CLIP model, augmented with learnable prompts to process multi-level features within a Soldier-Officer Window Self-Attention (SOWA) framework. Our method has been rigorously evaluated on five benchmark datasets, achieving superior performance by leading in 18 out of 20 metrics, setting a new standard against existing state-of-the-art techniques.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Chart understanding and anomaly detection, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 33 of its 43 references that are in the map. First posted on arXiv 4 Jul 2024.

Builds on

33 of 43 references.

References · 33

  1. 2017Attention Is All You Needcited by 60,361
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  5. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,915
  6. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,864
  7. 2016V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentationcited by 2,292
  8. 2022Conditional Prompt Learning for Vision-Language Modelscited by 1,433
  9. 2021Open-vocabulary Object Detection via Vision and Language Knowledge Distillationcited by 906
  10. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 776
  11. 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 594
  12. 2021Extract Free Dense Labels from CLIPcited by 583
21 more
  1. 2021DenseCLIP: Language-Guided Dense Prediction with Context-Aware Promptingcited by 532
  2. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 531
  3. 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 514
  4. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 421
  5. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
  6. 2019Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddingscited by 386
  7. 2021CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flowscited by 321
  8. 2020Sub-Image Anomaly Detection with Deep Pyramid Correspondencescited by 300
  9. 2023AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detectioncited by 271
  10. 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
  11. 2020Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentationcited by 212
  12. 2023EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latenciescited by 160
  13. 2022Registration based Few-Shot Anomaly Detectioncited by 133
  14. 2023APRIL-GAN: A Zero-/Few-Shot Anomaly Classification and Segmentation Method for CVPR 2023 VAND Workshop Challenge Tracks 1&2: 1st Place on Zero-shot AD and 4th Place on Few-shot ADcited by 130
  15. 2021Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detectioncited by 119
  16. 2022DualCoOp: Fast Adaptation to Multi-Label Recognition with Limited Annotationscited by 102
  17. 2024PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detectioncited by 98
  18. 2023A Closer Look at the Explainability of Contrastive Language-Image Pre-trainingcited by 96
  19. 2024Toward Generalist Anomaly Detection via In-context Residual Learning with Few-shot Sample Promptscited by 89
  20. 2022PNI : Industrial Anomaly Detection using Position and Neighborhood Informationcited by 69
  21. 2021A Hierarchical Transformation-Discriminating Generative Model for Few Shot Anomaly Detectioncited by 60

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 1.

2 papers.

  1. 2025A Survey on Foundation-Model-Based Industrial Defect Detectioncited by 5
  2. 2026PostureObjectstitch: Anomaly Image Generation Considering Assembly Relationships in Industrial Scenarioscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Anomaly Detection by Adapting a pre-trained Vision Language Model
  2. 2026SSVP: Synergistic Semantic-Visual Prompting for Industrial Zero-Shot Anomaly Detection
  3. 2024FADE: Few-shot/zero-shot Anomaly Detection Engine using Large Vision-Language Model
  4. 2024VMAD: Visual-enhanced Multimodal Large Language Model for Zero-Shot Anomaly Detection
  5. 2026A Structured Benchmark for Text-Guided Anomaly Detection: When Language Stops Conditioning the Decision
  6. 2025Detect, Classify, Act: Categorizing Industrial Anomalies with Multi-Modal Large Language Models
  7. 2025SAGE: A Visual Language Model for Anomaly Detection via Fact Enhancement and Entropy-aware Alignment
  8. 2023AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Models