{"url":"/task/anomaly-classification","name":"Anomaly Classification","slug":"anomaly-classification","description_markdown":"Anomaly Classification is the task of identifying and categorizing different types of anomalies in visual data, rather than simply detecting whether an input is normal or anomalous. Unlike anomaly detection, which is typically a binary classification (normal vs. anomaly), anomaly classification requires distinguishing between multiple anomaly classes—each representing a distinct type of anomaly or irregularity. This task is critical in real-world applications such as industrial inspection, where different anomalies may require different responses or interventions.","categories":[{"name":"Computer Vision","url":"/area/computer-vision"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":72,"papers_with_code":33,"benchmarks":5,"benchmark_tables_in_archive":5,"benchmark_tables_shown":5,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":9,"subtasks":1,"parent_tasks":2},"benchmarks":[{"leaderboard":"/sota/anomaly-classification-on-goodsad","slug":"anomaly-classification-on-goodsad","dataset":"GoodsAD","dataset_url":"/dataset/goodsad","rows_in_archive":11,"metrics":["AUPR","AUROC"],"first_row_in_archive_order":{"model":"PatchCore-100%","paper_title":"Towards Total Recall in Industrial Anomaly Detection","paper_url":"/paper/towards-total-recall-in-industrial-anomaly","paper_date":"2021-06-15","arxiv_id":"2106.08265","code_links":[{"title":"openvinotoolkit/anomalib","url":"https://github.com/openvinotoolkit/anomalib"},{"title":"amazon-science/patchcore-inspection","url":"https://github.com/amazon-science/patchcore-inspection"},{"title":"amazon-research/patchcore-inspection","url":"https://github.com/amazon-research/patchcore-inspection"},{"title":"hcw-00/PatchCore_anomaly_detection","url":"https://github.com/hcw-00/PatchCore_anomaly_detection"},{"title":"mindspore-ai/models","url":"https://github.com/mindspore-ai/models/tree/master/official/cv/patchcore"},{"title":"rvorias/ind_knn_ad","url":"https://github.com/rvorias/ind_knn_ad"},{"title":"OpenAOI/anodet","url":"https://github.com/OpenAOI/anodet"},{"title":"Burf/tfdetection","url":"https://github.com/Burf/tfdetection"},{"title":"tbcvContributor/DeepHawkeye","url":"https://github.com/tbcvContributor/DeepHawkeye"},{"title":"Ultranity/Anomaly.Paddle","url":"https://github.com/Ultranity/Anomaly.Paddle"},{"title":"tiskw/patchcore-ad","url":"https://github.com/tiskw/patchcore-ad"},{"title":"any-tech/PatchCore-ex","url":"https://github.com/any-tech/PatchCore-ex"},{"title":"taikiinoue45/PatchCore","url":"https://github.com/taikiinoue45/PatchCore"},{"title":"JoegameZhou/PatchCore","url":"https://github.com/JoegameZhou/PatchCore"},{"title":"captainfffsama/pathcore","url":"https://github.com/captainfffsama/pathcore"},{"title":"yangyucheng000/patchcore","url":"https://github.com/yangyucheng000/patchcore"},{"title":"totoroKalic/patchcore-mindspore","url":"https://github.com/totoroKalic/patchcore-mindspore"},{"title":"2023-MindSpore-1/ms-code-4","url":"https://github.com/2023-MindSpore-1/ms-code-4/tree/main/PatchCore"}],"syntology":{"n":36,"n_ran":5,"n_unverified":31,"n_pointer_only":0}}},{"leaderboard":"/sota/anomaly-classification-on-mvtecad","slug":"anomaly-classification-on-mvtecad","dataset":"MVTecAD","dataset_url":"/dataset/mvtecad","rows_in_archive":2,"metrics":["Accuracy (% )"],"first_row_in_archive_order":{"model":"VELM","paper_title":"Detect, Classify, Act: Categorizing Industrial Anomalies with Multi-Modal Large Language Models","paper_url":"/paper/detect-classify-act-categorizing-industrial","paper_date":"2025-05-05","arxiv_id":"2505.02626","code_links":[{"title":"sassanmtr/velm","url":"https://github.com/sassanmtr/velm"}],"syntology":null}},{"leaderboard":"/sota/anomaly-classification-on-mvtec-ac","slug":"anomaly-classification-on-mvtec-ac","dataset":"MVTec-AC","dataset_url":"/dataset/mvtec-ac","rows_in_archive":1,"metrics":["Accuracy (% )"],"first_row_in_archive_order":{"model":"VELM","paper_title":"Detect, Classify, Act: Categorizing Industrial Anomalies with Multi-Modal Large Language Models","paper_url":"/paper/detect-classify-act-categorizing-industrial","paper_date":"2025-05-05","arxiv_id":"2505.02626","code_links":[{"title":"sassanmtr/velm","url":"https://github.com/sassanmtr/velm"}],"syntology":null}},{"leaderboard":"/sota/anomaly-classification-on-visa","slug":"anomaly-classification-on-visa","dataset":"VisA","dataset_url":"/dataset/visa","rows_in_archive":1,"metrics":["Detection AUROC"],"first_row_in_archive_order":{"model":"APRIL-GAN","paper_title":"APRIL-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 AD","paper_url":"/paper/a-zero-few-shot-anomaly-classification-and","paper_date":"2023-05-27","arxiv_id":"2305.17382","code_links":[{"title":"bychelsea/vand-april-gan","url":"https://github.com/bychelsea/vand-april-gan"},{"title":"hq-deng/AnoVL","url":"https://github.com/hq-deng/AnoVL"}],"syntology":{"n":10,"n_ran":4,"n_unverified":6,"n_pointer_only":0}}},{"leaderboard":"/sota/anomaly-classification-on-visa-ac","slug":"anomaly-classification-on-visa-ac","dataset":"VisA-AC","dataset_url":"/dataset/visa-ac","rows_in_archive":1,"metrics":["Accuracy(%)"],"first_row_in_archive_order":{"model":"VELM","paper_title":"Detect, Classify, Act: Categorizing Industrial Anomalies with Multi-Modal Large Language Models","paper_url":"/paper/detect-classify-act-categorizing-industrial","paper_date":"2025-05-05","arxiv_id":"2505.02626","code_links":[{"title":"sassanmtr/velm","url":"https://github.com/sassanmtr/velm"}],"syntology":null}}],"datasets":[{"url":"/dataset/mvtecad","name":"MVTecAD","full_name":"MVTEC ANOMALY DETECTION DATASET","num_papers_in_archive":402},{"url":"/dataset/visa","name":"VisA","full_name":"Visual Anomaly Dataset","num_papers_in_archive":86},{"url":"/dataset/goodsad","name":"GoodsAD","full_name":"","num_papers_in_archive":15},{"url":"/dataset/elpv","name":"ELPV","full_name":"A dataset of functional and defective solar cells extracted from EL images of solar modules","num_papers_in_archive":14},{"url":"/dataset/edeniss2020","name":"edeniss2020","full_name":"EDEN ISS 2020 Telemetry Dataset","num_papers_in_archive":2},{"url":"/dataset/hawk-annotation-dataset","name":"Hawk Annotation Dataset","full_name":"","num_papers_in_archive":1},{"url":"/dataset/mvtec-ac","name":"MVTec-AC","full_name":"","num_papers_in_archive":1},{"url":"/dataset/visa-ac","name":"VisA-AC","full_name":"","num_papers_in_archive":1},{"url":"/dataset/lusitano-fabric-defect-detection-dataset","name":"Lusitano Fabric Defect Detection Dataset","full_name":"Lusitano Fabric Defect Detection Dataset","num_papers_in_archive":0}],"subtasks":[{"url":"/task/anomaly-severity-classification-anomaly-vs","name":"Anomaly Severity Classification (Anomaly vs. Defect)"}],"parent_tasks":[{"url":"/task/2d-classification","name":"2D Classification"},{"url":"/task/anomaly-detection","name":"Anomaly Detection"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":33,"tagged_in_all":72,"items":[{"url":"/paper/towards-total-recall-in-industrial-anomaly","title":"Towards Total Recall in Industrial Anomaly Detection","date":"2021-06-15","arxiv_id":"2106.08265","repositories_listed":18,"syntology":{"n":36,"n_ran":5,"n_unverified":31,"n_pointer_only":0}},{"url":"/paper/anomaly-detection-via-reverse-distillation","title":"Anomaly Detection via Reverse Distillation from One-Class Embedding","date":"2022-01-26","arxiv_id":"2201.10703","repositories_listed":5,"syntology":{"n":18,"n_ran":14,"n_unverified":4,"n_pointer_only":15}},{"url":"/paper/sub-image-anomaly-detection-with-deep-pyramid","title":"Sub-Image Anomaly Detection with Deep Pyramid Correspondences","date":"2020-05-05","arxiv_id":"2005.02357","repositories_listed":5,"syntology":{"n":3,"n_ran":0,"n_unverified":3,"n_pointer_only":0}},{"url":"/paper/winclip-zero-few-shot-anomaly-classification","title":"WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation","date":"2023-03-26","arxiv_id":"2303.14814","repositories_listed":4,"syntology":{"n":12,"n_ran":5,"n_unverified":7,"n_pointer_only":0}},{"url":"/paper/self-supervised-out-of-distribution-detection-1","title":"Natural Synthetic Anomalies for Self-Supervised Anomaly Detection and Localization","date":"2021-09-30","arxiv_id":"2109.15222","repositories_listed":3,"syntology":{"n":8,"n_ran":2,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/draem-a-discriminatively-trained","title":"DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection","date":"2021-08-17","arxiv_id":"2108.07610","repositories_listed":3,"syntology":{"n":7,"n_ran":2,"n_unverified":5,"n_pointer_only":0}},{"url":"/paper/cflow-ad-real-time-unsupervised-anomaly","title":"CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flows","date":"2021-07-27","arxiv_id":"2107.12571","repositories_listed":3,"syntology":{"n":12,"n_ran":3,"n_unverified":9,"n_pointer_only":0}},{"url":"/paper/binet-multi-perspective-business-process","title":"BINet: Multi-perspective Business Process Anomaly Classification","date":"2019-02-08","arxiv_id":"1902.03155","repositories_listed":3,"syntology":null},{"url":"/paper/f-anogan-fast-unsupervised-anomaly-detection","title":"f-AnoGAN: Fast Unsupervised Anomaly Detection with Generative Adversarial Networks","date":"2019-01-30","arxiv_id":null,"repositories_listed":3,"syntology":null},{"url":"/paper/a-zero-few-shot-anomaly-classification-and","title":"APRIL-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 AD","date":"2023-05-27","arxiv_id":"2305.17382","repositories_listed":2,"syntology":{"n":10,"n_ran":4,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/simplenet-a-simple-network-for-image-anomaly","title":"SimpleNet: A Simple Network for Image Anomaly Detection and Localization","date":"2023-03-27","arxiv_id":"2303.15140","repositories_listed":2,"syntology":{"n":17,"n_ran":10,"n_unverified":7,"n_pointer_only":0}},{"url":"/paper/cutpaste-self-supervised-learning-for-anomaly","title":"CutPaste: Self-Supervised Learning for Anomaly Detection and Localization","date":"2021-04-08","arxiv_id":"2104.04015","repositories_listed":2,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/a-cytology-dataset-for-early-detection-of","title":"A Cytology Dataset for Early Detection of Oral Squamous Cell Carcinoma","date":"2025-06-11","arxiv_id":"2506.09661","repositories_listed":1,"syntology":null},{"url":"/paper/few-shot-anomaly-driven-generation-for","title":"Few-Shot Anomaly-Driven Generation for Anomaly Classification and Segmentation","date":"2025-05-14","arxiv_id":"2505.09263","repositories_listed":1,"syntology":null},{"url":"/paper/detect-classify-act-categorizing-industrial","title":"Detect, Classify, Act: Categorizing Industrial Anomalies with Multi-Modal Large Language Models","date":"2025-05-05","arxiv_id":"2505.02626","repositories_listed":1,"syntology":null},{"url":"/paper/video-anomaly-detection-with-structured","title":"Video Anomaly Detection with Structured Keywords","date":"2025-03-07","arxiv_id":"2503.10653","repositories_listed":1,"syntology":null},{"url":"/paper/anomaly-detection-in-cooperative-vehicle","title":"Anomaly Detection in Cooperative Vehicle Perception Systems under Imperfect Communication","date":"2025-01-28","arxiv_id":"2501.17329","repositories_listed":1,"syntology":null},{"url":"/paper/softpatch-fully-unsupervised-anomaly","title":"SoftPatch+: Fully Unsupervised Anomaly Classification and Segmentation","date":"2024-12-30","arxiv_id":"2412.20870","repositories_listed":1,"syntology":null},{"url":"/paper/multi-class-abnormality-classification-task","title":"Multi-Class Abnormality Classification Task in Video Capsule Endoscopy","date":"2024-10-25","arxiv_id":"2410.19973","repositories_listed":1,"syntology":null},{"url":"/paper/anomalyncd-towards-novel-anomaly-class","title":"AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios","date":"2024-10-18","arxiv_id":"2410.14379","repositories_listed":1,"syntology":null},{"url":"/paper/dualanodiff-dual-interrelated-diffusion-model","title":"DualAnoDiff: Dual-Interrelated Diffusion Model for Few-Shot Anomaly Image Generation","date":"2024-08-24","arxiv_id":"2408.13509","repositories_listed":1,"syntology":null},{"url":"/paper/unraveling-anomalies-in-time-unsupervised","title":"Unraveling Anomalies in Time: Unsupervised Discovery and Isolation of Anomalous Behavior in Bio-regenerative Life Support System Telemetry","date":"2024-06-14","arxiv_id":"2406.09825","repositories_listed":1,"syntology":null},{"url":"/paper/minimaxad-a-lightweight-autoencoder-for","title":"MiniMaxAD: A Lightweight Autoencoder for Feature-Rich Anomaly Detection","date":"2024-05-16","arxiv_id":"2405.09933","repositories_listed":1,"syntology":null},{"url":"/paper/medianomaly-a-comparative-study-of-anomaly","title":"MedIAnomaly: A comparative study of anomaly detection in medical images","date":"2024-04-06","arxiv_id":"2404.04518","repositories_listed":1,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/adapting-visual-language-models-for","title":"Adapting Visual-Language Models for Generalizable Anomaly Detection in Medical Images","date":"2024-03-19","arxiv_id":"2403.12570","repositories_listed":1,"syntology":{"n":15,"n_ran":8,"n_unverified":7,"n_pointer_only":0}},{"url":"/paper/musc-zero-shot-industrial-anomaly","title":"MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled Images","date":"2024-01-30","arxiv_id":"2401.16753","repositories_listed":1,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/beyond-traditional-approaches-multi-task","title":"Beyond Traditional Approaches: Multi-Task Network for Breast Ultrasound Diagnosis","date":"2024-01-14","arxiv_id":"2401.07326","repositories_listed":1,"syntology":null},{"url":"/paper/multi-task-learning-for-joint-weakly","title":"Multi-task learning for joint weakly-supervised segmentation and aortic arch anomaly classification in fetal cardiac MRI","date":"2023-11-13","arxiv_id":"2311.07234","repositories_listed":1,"syntology":null},{"url":"/paper/a-prototype-based-neural-network-for-image","title":"A Prototype-Based Neural Network for Image Anomaly Detection and Localization","date":"2023-10-04","arxiv_id":"2310.02576","repositories_listed":1,"syntology":null},{"url":"/paper/pku-goodsad-a-supermarket-goods-dataset-for","title":"PKU-GoodsAD: A Supermarket Goods Dataset for Unsupervised Anomaly Detection and Segmentation","date":"2023-07-11","arxiv_id":"2307.04956","repositories_listed":1,"syntology":null}],"syntology_records":13,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}