{"url":"/sota/anomaly-detection-on-one-class-cifar-10","task":{"name":"Anomaly Detection","url":"/task/anomaly-detection","note":null},"dataset":{"name":"One-class CIFAR-10","url":"/dataset/cifar-10"},"category":"Computer Vision","categories":["Computer Vision","Graphs","Methodology","Miscellaneous"],"category_note":null,"description":"**Anomaly Detection** is a binary classification identifying unusual or unexpected patterns in a dataset, which deviate significantly from the majority of the data. The goal of anomaly detection is to identify such anomalies, which could represent errors, fraud, or other types of unusual events, and flag them for further investigation.\r\n\r\n[Image source]: [GAN-based Anomaly Detection in Imbalance Problems](https://paperswithcode.com/paper/gan-based-anomaly-detection-in-imbalance)","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["AUROC"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"AUROC":null}},"counts":{"rows":36,"rows_with_code":28,"rows_with_paper_page":36,"rows_dated":36,"rows_using_additional_data":9},"rows":[{"rank_in_archive_order":1,"model":"CLIP (OE)","metrics":{"AUROC":"99.6"},"uses_additional_data":true,"paper_date":"2022-05-23","paper":"/paper/exposing-outlier-exposure-what-can-be-learned","paper_url":"https://arxiv.org/abs/2205.11474v2","paper_title":"Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images","code":"https://github.com/liznerski/eoe","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"GeneralAD","metrics":{"AUROC":"99.3"},"uses_additional_data":false,"paper_date":"2024-07-17","paper":"/paper/generalad-anomaly-detection-across-domains-by","paper_url":"https://arxiv.org/abs/2407.12427v1","paper_title":"GeneralAD: Anomaly Detection Across Domains by Attending to Distorted Features","code":"https://github.com/LucStrater/GeneralAD","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":1,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"Fake It Till You Make It","metrics":{"AUROC":"99.1"},"uses_additional_data":true,"paper_date":"2022-05-28","paper":"/paper/fake-it-till-you-make-it-near-distribution","paper_url":"https://arxiv.org/abs/2205.14297v2","paper_title":"Fake It Till You Make It: Towards Accurate Near-Distribution Novelty Detection","code":"https://github.com/rohban-lab/fitymi","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"BLISS","metrics":{"AUROC":"99.1"},"uses_additional_data":false,"paper_date":"2024-07-24","paper":"/paper/when-text-and-images-don-t-mix-bias","paper_url":"https://arxiv.org/abs/2407.17083v1","paper_title":"When Text and Images Don't Mix: Bias-Correcting Language-Image Similarity Scores for Anomaly Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"PANDA-OE","metrics":{"AUROC":"98.9"},"uses_additional_data":true,"paper_date":"2020-10-12","paper":"/paper/panda-adapting-pretrained-features-for","paper_url":"https://arxiv.org/abs/2010.05903v3","paper_title":"PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation","code":"https://github.com/talreiss/PANDA","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"Mean-Shifted Contrastive Loss","metrics":{"AUROC":"98.6"},"uses_additional_data":true,"paper_date":"2021-06-07","paper":"/paper/mean-shifted-contrastive-loss-for-anomaly","paper_url":"https://arxiv.org/abs/2106.03844v2","paper_title":"Mean-Shifted Contrastive Loss for Anomaly Detection","code":"https://github.com/talreiss/Mean-Shifted-Anomaly-Detection","n_code_links":2,"syntology":null},{"rank_in_archive_order":7,"model":"CLIP (zero shot)","metrics":{"AUROC":"98.5"},"uses_additional_data":false,"paper_date":"2022-05-23","paper":"/paper/exposing-outlier-exposure-what-can-be-learned","paper_url":"https://arxiv.org/abs/2205.11474v2","paper_title":"Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images","code":"https://github.com/liznerski/eoe","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"DINO-FT","metrics":{"AUROC":"98.4"},"uses_additional_data":false,"paper_date":"2022-10-19","paper":"/paper/anomaly-detection-requires-better","paper_url":"https://arxiv.org/abs/2210.10773v1","paper_title":"Anomaly Detection Requires Better Representations","code":"https://github.com/eliahuhorwitz/3D-ADS","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"Transformaly","metrics":{"AUROC":"98.3"},"uses_additional_data":true,"paper_date":"2021-12-08","paper":"/paper/transformaly-two-feature-spaces-are-better","paper_url":"https://arxiv.org/abs/2112.04185v2","paper_title":"Transformaly -- Two (Feature Spaces) Are Better Than One","code":"https://github.com/MatanCohen1/Transformaly","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"CAP","metrics":{"AUROC":"97.0"},"uses_additional_data":false,"paper_date":"2021-12-05","paper":"/paper/simple-adaptive-projection-with-pretrained","paper_url":"https://arxiv.org/abs/2112.02597v2","paper_title":"Constrained Adaptive Projection with Pretrained Features for Anomaly Detection","code":"https://github.com/tabguigui/cap","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"PANDA","metrics":{"AUROC":"96.2"},"uses_additional_data":false,"paper_date":"2020-10-12","paper":"/paper/panda-adapting-pretrained-features-for","paper_url":"https://arxiv.org/abs/2010.05903v3","paper_title":"PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation","code":"https://github.com/talreiss/PANDA","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"CSI","metrics":{"AUROC":"94.3"},"uses_additional_data":false,"paper_date":"2020-07-16","paper":"/paper/csi-novelty-detection-via-contrastive","paper_url":"https://arxiv.org/abs/2007.08176v2","paper_title":"CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances","code":"https://github.com/alinlab/CSI","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":13,"model":"DUIAD","metrics":{"AUROC":"92.6"},"uses_additional_data":false,"paper_date":"2020-12-09","paper":"/paper/deep-unsupervised-image-anomaly-detection-an","paper_url":"https://arxiv.org/abs/2012.04837v1","paper_title":"Deep Unsupervised Image Anomaly Detection: An Information Theoretic Framework","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":14,"model":"DN2","metrics":{"AUROC":"92.5"},"uses_additional_data":true,"paper_date":"2020-02-24","paper":"/paper/deep-nearest-neighbor-anomaly-detection","paper_url":"https://arxiv.org/abs/2002.10445v1","paper_title":"Deep Nearest Neighbor Anomaly Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":15,"model":"DisAug CLR","metrics":{"AUROC":"92.5"},"uses_additional_data":false,"paper_date":"2020-11-04","paper":"/paper/learning-and-evaluating-representations-for-1","paper_url":"https://arxiv.org/abs/2011.02578v2","paper_title":"Learning and Evaluating Representations for Deep One-class Classification","code":"https://github.com/google-research/deep_representation_one_class","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"FCDD","metrics":{"AUROC":"92"},"uses_additional_data":true,"paper_date":"2020-07-03","paper":"/paper/explainable-deep-one-class-classification","paper_url":"https://arxiv.org/abs/2007.01760v3","paper_title":"Explainable Deep One-Class Classification","code":"https://github.com/liznerski/fcdd","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"IGD (pre-trained SSL)","metrics":{"AUROC":"91.25"},"uses_additional_data":false,"paper_date":"2021-01-25","paper":"/paper/unsupervised-anomaly-detection-and","paper_url":"https://arxiv.org/abs/2101.10043v5","paper_title":"Deep One-Class Classification via Interpolated Gaussian Descriptor","code":"https://github.com/tianyu0207/IGD","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":18,"model":"GAN based Anomaly Detection in Imbalance Problems","metrics":{"AUROC":"90.6"},"uses_additional_data":true,"paper_date":"2020-08-28","paper":"/paper/gan-based-anomaly-detection-in-imbalance","paper_url":"http://intlab.skuniv.ac.kr/paper/GAN-based_Anomaly_Detection_in_Imbalance_Problems.pdf","paper_title":"GAN-based Anomaly Detection in Imbalance Problems","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":19,"model":"SSOOD","metrics":{"AUROC":"90.1"},"uses_additional_data":false,"paper_date":"2019-06-28","paper":"/paper/using-self-supervised-learning-can-improve","paper_url":"https://arxiv.org/abs/1906.12340v2","paper_title":"Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty","code":"https://github.com/hendrycks/ss-ood","n_code_links":4,"syntology":{"n_ran":5,"n_unverified":7,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":20,"model":"SSD","metrics":{"AUROC":"90.0"},"uses_additional_data":false,"paper_date":"2021-03-22","paper":"/paper/ssd-a-unified-framework-for-self-supervised-1","paper_url":"https://arxiv.org/abs/2103.12051v1","paper_title":"SSD: A Unified Framework for Self-Supervised Outlier Detection","code":"https://github.com/inspire-group/SSD","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":21,"model":"UTAD","metrics":{"AUROC":"88.4"},"uses_additional_data":false,"paper_date":"2021-03-22","paper":"/paper/unsupervised-two-stage-anomaly-detection","paper_url":"https://arxiv.org/abs/2103.11671v1","paper_title":"Unsupervised Two-Stage Anomaly Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":22,"model":"GOAD","metrics":{"AUROC":"88.2"},"uses_additional_data":false,"paper_date":"2020-05-05","paper":"/paper/classification-based-anomaly-detection-for-1","paper_url":"https://arxiv.org/abs/2005.02359v1","paper_title":"Classification-Based Anomaly Detection for General Data","code":"https://github.com/xuhongzuo/DeepOD","n_code_links":2,"syntology":{"n_ran":5,"n_unverified":5,"n_samples":10,"n_pointer_only_licence":4}},{"rank_in_archive_order":23,"model":"ARNET","metrics":{"AUROC":"86.6"},"uses_additional_data":false,"paper_date":"2019-11-25","paper":"/paper/inverse-transform-autoencoder-for-anomaly","paper_url":"https://arxiv.org/abs/1911.10676v3","paper_title":"Attribute Restoration Framework for Anomaly Detection","code":"https://github.com/FishSmile-syx/ITAE-Pytorch-Anomaly_Detection","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"Reverse Distillation","metrics":{"AUROC":"86.5"},"uses_additional_data":false,"paper_date":"2022-01-26","paper":"/paper/anomaly-detection-via-reverse-distillation","paper_url":"https://arxiv.org/abs/2201.10703v2","paper_title":"Anomaly Detection via Reverse Distillation from One-Class Embedding","code":"https://github.com/openvinotoolkit/anomalib/tree/development/anomalib/models/reverse_distillation","n_code_links":5,"syntology":{"n_ran":14,"n_unverified":4,"n_samples":18,"n_pointer_only_licence":15}},{"rank_in_archive_order":25,"model":"ADT","metrics":{"AUROC":"86"},"uses_additional_data":false,"paper_date":"2018-05-28","paper":"/paper/deep-anomaly-detection-using-geometric","paper_url":"http://arxiv.org/abs/1805.10917v2","paper_title":"Deep Anomaly Detection Using Geometric Transformations","code":"https://github.com/izikgo/AnomalyDetectionTransformations","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":5,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"IGD (pre-trained ImageNet)","metrics":{"AUROC":"83.68"},"uses_additional_data":false,"paper_date":"2021-01-25","paper":"/paper/unsupervised-anomaly-detection-and","paper_url":"https://arxiv.org/abs/2101.10043v5","paper_title":"Deep One-Class Classification via Interpolated Gaussian Descriptor","code":"https://github.com/tianyu0207/IGD","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":27,"model":"ESAD","metrics":{"AUROC":"83.3"},"uses_additional_data":true,"paper_date":"2020-12-09","paper":"/paper/esad-end-to-end-deep-semi-supervised-anomaly","paper_url":"https://arxiv.org/abs/2012.04905v3","paper_title":"ESAD: End-to-end Deep Semi-supervised Anomaly Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":28,"model":"IGD (scratch)","metrics":{"AUROC":"74.33"},"uses_additional_data":false,"paper_date":"2021-01-25","paper":"/paper/unsupervised-anomaly-detection-and","paper_url":"https://arxiv.org/abs/2101.10043v5","paper_title":"Deep One-Class Classification via Interpolated Gaussian Descriptor","code":"https://github.com/tianyu0207/IGD","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":29,"model":"P-KDGAN","metrics":{"AUROC":"73.05"},"uses_additional_data":false,"paper_date":"2020-07-14","paper":"/paper/p-kdgan-progressive-knowledge-distillation","paper_url":"https://arxiv.org/abs/2007.06963v2","paper_title":"P-KDGAN: Progressive Knowledge Distillation with GANs for One-class Novelty Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":30,"model":"OLED","metrics":{"AUROC":"67.1"},"uses_additional_data":false,"paper_date":"2021-03-27","paper":"/paper/oled-one-class-learned-encoder-decoder","paper_url":"https://arxiv.org/abs/2103.14953v3","paper_title":"OLED: One-Class Learned Encoder-Decoder Network with Adversarial Context Masking for Novelty Detection","code":"https://github.com/jewelltaylor/OLED","n_code_links":1,"syntology":null},{"rank_in_archive_order":31,"model":"OCGAN","metrics":{"AUROC":"66.83"},"uses_additional_data":false,"paper_date":"2019-03-20","paper":"/paper/ocgan-one-class-novelty-detection-using-gans","paper_url":"http://arxiv.org/abs/1903.08550v1","paper_title":"OCGAN: One-class Novelty Detection Using GANs with Constrained Latent Representations","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":32,"model":"FastFlow","metrics":{"AUROC":"66.7"},"uses_additional_data":false,"paper_date":"2021-11-15","paper":"/paper/fastflow-unsupervised-anomaly-detection-and","paper_url":"https://arxiv.org/abs/2111.07677v2","paper_title":"FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows","code":"https://github.com/openvinotoolkit/anomalib/tree/main/src/anomalib/models/fastflow","n_code_links":5,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":33,"model":"DASVDD","metrics":{"AUROC":"66.5"},"uses_additional_data":false,"paper_date":"2021-06-09","paper":"/paper/dasvdd-deep-autoencoding-support-vector-data","paper_url":"https://arxiv.org/abs/2106.05410v4","paper_title":"DASVDD: Deep Autoencoding Support Vector Data Descriptor for Anomaly Detection","code":"https://github.com/Armanfard-Lab/DASVDD","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":34,"model":"Deep SVDD","metrics":{"AUROC":"65.7"},"uses_additional_data":false,"paper_date":"2018-07-01","paper":"/paper/deep-one-class-classification","paper_url":"https://icml.cc/Conferences/2018/Schedule?showEvent=2483","paper_title":"Deep One-Class Classification","code":"https://github.com/lukasruff/Deep-SVDD-PyTorch","n_code_links":1,"syntology":null},{"rank_in_archive_order":35,"model":"Self-Supervised DeepSVDD","metrics":{"AUROC":"64.8"},"uses_additional_data":false,"paper_date":"2020-10-12","paper":"/paper/panda-adapting-pretrained-features-for","paper_url":"https://arxiv.org/abs/2010.05903v3","paper_title":"PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation","code":"https://github.com/talreiss/PANDA","n_code_links":1,"syntology":null},{"rank_in_archive_order":36,"model":"Self-Supervised One-class SVM, RBF kernel","metrics":{"AUROC":"64.7"},"uses_additional_data":false,"paper_date":"2020-10-12","paper":"/paper/panda-adapting-pretrained-features-for","paper_url":"https://arxiv.org/abs/2010.05903v3","paper_title":"PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation","code":"https://github.com/talreiss/PANDA","n_code_links":1,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":15,"rows_with_any_sample_ran":10,"distinct_papers_with_graph_line":12,"distinct_papers_with_any_sample_ran":8,"samples_over_distinct_papers":{"n_ran":32,"n_unverified":30,"n_samples":62,"n_pointer_only_licence":23,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":34,"n_unverified":31,"n_samples":65,"n_pointer_only_licence":25,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}