mlsanityData to 14 Oct 2026

AA-CLIP: Enhancing Zero-shot Anomaly Detection via Anomaly-Aware CLIP

arXiv 2503.06661 ↗First posted 9 Mar 2025cs.CV cs.AIcited by 73

Not on the map. Why?

Abstract (arXiv)

Anomaly detection (AD) identifies outliers for applications like defect and lesion detection. While CLIP shows promise for zero-shot AD tasks due to its strong generalization capabilities, its inherent Anomaly-Unawareness leads to limited discrimination between normal and abnormal features. To address this problem, we propose Anomaly-Aware CLIP (AA-CLIP), which enhances CLIP's anomaly discrimination ability in both text and visual spaces while preserving its generalization capability. AA-CLIP is achieved through a straightforward yet effective two-stage approach: it first creates anomaly-aware text anchors to differentiate normal and abnormal semantics clearly, then aligns patch-level visual features with these anchors for precise anomaly localization. This two-stage strategy, with the help of residual adapters, gradually adapts CLIP in a controlled manner, achieving effective AD while maintaining CLIP's class knowledge. Extensive experiments validate AA-CLIP as a resource-efficient solution for zero-shot AD tasks, achieving state-of-the-art results in industrial and medical applications. The code is available at https://github.com/Mwxinnn/AA-CLIP.

Key facts. Cited by 73 papers in the map (data through 14 Oct 2026). Builds on 51 of its 64 references that are in the map. First posted on arXiv 9 Mar 2025.

Builds on

51 of 64 references.

References · 51

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  3. 2022BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generationcited by 4,322
  4. 2021Align before Fuse: Vision and Language Representation Learning with Momentum Distillationcited by 1,619
  5. 2021CLIP-Adapter: Better Vision-Language Models with Feature Adapterscited by 1,026
  6. 2021StyleCLIP: Text-Driven Manipulation of StyleGAN Imagerycited by 991
  7. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 776
  8. 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 594
  9. 2017Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discoverycited by 566
  10. 2021ClipCap: CLIP Prefix for Image Captioningcited by 559
  11. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 421
  12. 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 400
38 more
  1. 2022Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classificationcited by 362
  2. 2021CRIS: CLIP-Driven Referring Image Segmentationcited by 353
  3. 2021CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flowscited by 321
  4. 2023WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentationcited by 315
  5. 2023Side Adapter Network for Open-Vocabulary Semantic Segmentationcited by 300
  6. 2023AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detectioncited by 271
  7. 2022A Unified Model for Multi-class Anomaly Detectioncited by 268
  8. 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
  9. 2021CLIP2Video: Mastering Video-Text Retrieval via Image CLIPcited by 225
  10. 2023CLIP-Driven Universal Model for Organ Segmentation and Tumor Detectioncited by 224
  11. 2021VT-ADL: A Vision Transformer Network for Image Anomaly Detection and Localizationcited by 223
  12. 2016A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Imagescited by 221
  13. 2023Teaching CLIP to Count to Tencited by 198
  14. 2022ZegCLIP: Towards Adapting CLIP for Zero-shot Semantic Segmentationcited by 178
  15. 2022DeSTSeg: Segmentation Guided Denoising Student-Teacher for Anomaly Detectioncited by 154
  16. 2021Student-Teacher Feature Pyramid Matching for Anomaly Detectioncited by 146
  17. 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
  18. 2020Deep Learning for Medical Anomaly Detection -- A Surveycited by 127
  19. 2024AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detectioncited by 127
  20. 2022ReCLIP: A Strong Zero-Shot Baseline for Referring Expression Comprehensioncited by 120
  21. 2022CLIP is Also an Efficient Segmenter: A Text-Driven Approach for Weakly Supervised Semantic Segmentationcited by 118
  22. 2023Diffusion Model as Representation Learnercited by 95
  23. 2023CLIP-Count: Towards Text-Guided Zero-Shot Object Countingcited by 83
  24. 2023Verbs in Action: Improving verb understanding in video-language modelscited by 76
  25. 2022Fine-grained Image Captioning with CLIP Rewardcited by 69
  26. 2023GraphAdapter: Tuning Vision-Language Models With Dual Knowledge Graphcited by 69
  27. 2024Adapting Visual-Language Models for Generalizable Anomaly Detection in Medical Imagescited by 69
  28. 2023Hierarchical Vector Quantized Transformer for Multi-class Unsupervised Anomaly Detectioncited by 66
  29. 2023Segment Any Anomaly without Training via Hybrid Prompt Regularizationcited by 60
  30. 2023CLIP-AD: A Language-Guided Staged Dual-Path Model for Zero-shot Anomaly Detectioncited by 58
  31. 2024VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentationcited by 50
  32. 2020Label-Free Segmentation of COVID-19 Lesions in Lung CTcited by 37
  33. 2023DiAD: A Diffusion-based Framework for Multi-class Anomaly Detectioncited by 33
  34. 2022ADTR: Anomaly Detection Transformer with Feature Reconstructioncited by 31
  35. 2022When are Lemons Purple? The Concept Association Bias of Vision-Language Modelscited by 23
  36. 2024DetailCLIP: Detail-Oriented CLIP for Fine-Grained Taskscited by 15
  37. 2025Towards Accurate Unified Anomaly Segmentationcited by 7
  38. 2024SRL-CLIP: Efficient CLIP Video Adaptation via Structured Semantic Role Labelscited by 4

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 20; 2026: 53.

73 papers.

Sort:

By area:

  1. 2025OmniAD: Detect and Understand Industrial Anomaly via Multimodal Reasoningcited by 18
  2. 2026Benchmarking Reward Hack Detection in Code Environments via Contrastive Analysiscited by 18
  3. 2026VisualAD: Language-Free Zero-Shot Anomaly Detection via Vision Transformercited by 11
  4. 2025AF-CLIP: Zero-Shot Anomaly Detection via Anomaly-Focused CLIP Adaptationcited by 7
  5. 2026MRAD: Zero-Shot Anomaly Detection with Memory-Driven Retrievalcited by 6
  6. 2026One Language-Free Foundation Model Is Enough for Universal Vision Anomaly Detectioncited by 4
  7. 2026AnomalyVFM -- Transforming Vision Foundation Models into Zero-Shot Anomaly Detectorscited by 4
  8. 2026GS-CLIP: Zero-shot 3D Anomaly Detection by Geometry-Aware Prompt and Synergistic View Representation Learningcited by 4
  9. 2025Crane: Context-Guided Prompt Learning and Attention Refinement for Zero-Shot Anomaly Detectioncited by 3
  10. 2025GenCLIP: Generalizing CLIP Prompts for Zero-shot Anomaly Detectioncited by 3
  11. 2026Latent Anomaly Knowledge Excavation: Unveiling Sparse Sensitive Neurons in Vision-Language Modelscited by 3
  12. 2025A Comprehensive Survey for Real-World Industrial Defect Detection: Challenges, Approaches, and Prospectscited by 2
38 more
  1. 2025Quality-Aware Language-Conditioned Local Auto-Regressive Anomaly Synthesis and Detectioncited by 2
  2. 2025ADNet: A Large-Scale and Extensible Multi-Domain Benchmark for Anomaly Detection Across 380 Real-World Categoriescited by 2
  3. 2026Back to Point: Exploring Point-Language Models for Zero-Shot 3D Anomaly Detectioncited by 2
  4. 2026VitaTouch: Property-Aware Vision-Tactile-Language Model for Robotic Quality Inspection in Manufacturingcited by 2
  5. 2025IQE-CLIP: Instance-aware Query Embedding for Zero-/Few-shot Anomaly Detection in Medical Domaincited by 1
  6. 2025CoPS: Conditional Prompt Synthesis for Zero-Shot Anomaly Detectioncited by 1
  7. 2025TRACES: Temporal Recall with Contextual Embeddings for Real-Time Video Anomaly Detectioncited by 1
  8. 2025ABounD: Adversarial Boundary-Driven Few-Shot Learning for Multi-Class Anomaly Detectioncited by 1
  9. 2026HAAF: Hierarchical Adaptation and Alignment of Foundation Models for Few-Shot Pathology Anomaly Detectioncited by 1
  10. 2026DMP-3DAD: Cross-Category 3D Anomaly Detection via Realistic Depth Map Projection with Few Normal Samplescited by 1
  11. 2026AG-VAS: Anchor-Guided Zero-Shot Visual Anomaly Segmentation with Large Multimodal Modelscited by 1
  12. 2026WMoE-CLIP: Wavelet-Enhanced Mixture-of-Experts Prompt Learning for Zero-Shot Anomaly Detectioncited by 1
  13. 2026FB-CLIP: Fine-Grained Zero-Shot Anomaly Detection with Foreground-Background Disentanglementcited by 1
  14. 2026SPG: Sparse-Projected Guides with Sparse Autoencoders for Zero-Shot Anomaly Detectioncited by 1
  15. 2026PASTA: Vision Transformer Patch Aggregation for Weakly Supervised Target and Anomaly Segmentationcited by 1
  16. 2026AnomalyAgent: Training-Free Agentic Models for Zero-/Few-Shot Anomaly Detectioncited by 1
  17. 2025SD-MAD: Sign-Driven Few-shot Multi-Anomaly Detection in Medical Imagescited by 0
  18. 2025ViP2-CLIP: Visual-Perception Prompting with Unified Alignment for Zero-Shot Anomaly Detectioncited by 0
  19. 2025TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bankcited by 0
  20. 2025Unsupervised Integrated-Circuit Defect Segmentation via Image-Intrinsic Normalitycited by 0
  21. 2025Advancing Metallic Surface Defect Detection via Anomaly-Guided Pretraining on a Large Industrial Datasetcited by 0
  22. 2025TokenCLIP: Token-wise Prompt Learning for Zero-shot Anomaly Detectioncited by 0
  23. 2025Image-Intrinsic Priors for Integrated Circuit Defect Detection and Novel Class Discovery via Self-Supervised Learningcited by 0
  24. 2025ProtoAnomalyNCD: Prototype Learning for Multi-class Novel Anomaly Discovery in Industrial Scenarioscited by 0
  25. 2025AnoRefiner: Anomaly-Aware Group-Wise Refinement for Zero-Shot Industrial Anomaly Detectioncited by 0
  26. 2026HarmoniAD: Harmonizing Local Structures and Global Semantics for Anomaly Detectioncited by 0
  27. 2026Conditional Compatibility Learning for Context-Dependent Anomaly Detectioncited by 0
  28. 2026MoECLIP: Patch-Specialized Experts for Zero-shot Anomaly Detectioncited by 0
  29. 2026MedSAD-CLIP: Supervised CLIP with Token-Patch Cross-Attention for Medical Anomaly Detection and Segmentationcited by 0
  30. 2026SteelDefectX: A Multi-Form Vision-Language Dataset and Benchmark for Steel Surface Defect Analysiscited by 0
  31. 2026MApLe: Multi-instance Alignment of Diagnostic Reports and Large Medical Imagescited by 0
  32. 2026H2VLR: Heterogeneous Hypergraph Vision-Language Reasoning for Few-Shot Anomaly Detectioncited by 0
  33. 2026ZSG-IAD: A Multimodal Framework for Zero-Shot Grounded Industrial Anomaly Detectioncited by 0
  34. 2026EV-CLIP: Efficient Visual Prompt Adaptation for CLIP in Few-shot Action Recognition under Visual Challengescited by 0
  35. 2026Text-Guided Multimodal Unified Industrial Anomaly Detectioncited by 0
  36. 2026Align3D-AD: Cross-Modal Feature Alignment and Dual-Prompt Learning for Zero-shot 3D Anomaly Detectioncited by 0
  37. 2026MTL-MAD: Multi-Task Learners are Effective Medical Anomaly Detectorscited by 0
  38. 2026IPAD-CLIP: Teaching CLIP to Detect Image Local Perceptual Artifactscited by 0

50 most cited of 73 All 73 (JSON)