{"url":"/sota/multi-label-classification-on-pascal-voc-2007","task":{"name":"Multi-Label Classification","url":"/task/multi-label-classification","note":null},"dataset":{"name":"PASCAL VOC 2007","url":"/dataset/pascal-voc-2007"},"category":"Computer Vision","categories":["Computer Vision","Medical","Methodology","Reasoning"],"category_note":null,"description":"**Multi-Label Classification** is the supervised learning problem where an instance may be associated with multiple labels. This is an extension of single-label classification (i.e., multi-class, or binary) where each instance is only associated with a single class label.\r\n\r\n\r\n<span class=\"description-source\">Source: [Deep Learning for Multi-label Classification ](https://arxiv.org/abs/1502.05988)</span>","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":["mAP"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"mAP":"higher"}},"counts":{"rows":17,"rows_with_code":16,"rows_with_paper_page":17,"rows_dated":17,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"Q2L-CvT(ImageNet-21K pretrained, resolution 384)","metrics":{"mAP":"97.3"},"uses_additional_data":false,"paper_date":"2021-07-22","paper":"/paper/query2label-a-simple-transformer-way-to-multi","paper_url":"https://arxiv.org/abs/2107.10834v1","paper_title":"Query2Label: A Simple Transformer Way to Multi-Label Classification","code":"https://github.com/SlongLiu/query2labels","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"Q2L-TResL(ImageNet-21K pretrained, resolution 448)","metrics":{"mAP":"96.9"},"uses_additional_data":false,"paper_date":"2021-07-22","paper":"/paper/query2label-a-simple-transformer-way-to-multi","paper_url":"https://arxiv.org/abs/2107.10834v1","paper_title":"Query2Label: A Simple Transformer Way to Multi-Label Classification","code":"https://github.com/SlongLiu/query2labels","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"GKGNet","metrics":{"mAP":"96.8"},"uses_additional_data":false,"paper_date":"2023-08-28","paper":"/paper/gkgnet-group-k-nearest-neighbor-based-graph","paper_url":"https://arxiv.org/abs/2308.14378v3","paper_title":"GKGNet: Group K-Nearest Neighbor based Graph Convolutional Network for Multi-Label Image Recognition","code":"https://github.com/jin-s13/gkgnet","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"MLD-TResNetL-AAM (resolution 448, pretrain from OpenImages V6)","metrics":{"mAP":"96.70"},"uses_additional_data":false,"paper_date":"2022-09-14","paper":"/paper/combining-metric-learning-and-attention-heads","paper_url":"https://arxiv.org/abs/2209.06585v2","paper_title":"Combining Metric Learning and Attention Heads For Accurate and Efficient Multilabel Image Classification","code":"https://github.com/openvinotoolkit/deep-object-reid","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"M3TR(448×448)","metrics":{"mAP":"96.5"},"uses_additional_data":false,"paper_date":"2021-10-01","paper":"/paper/m3tr-multi-modal-multi-label-recognition-with","paper_url":"https://dl.acm.org/doi/10.1145/3474085.3475191","paper_title":"M3TR: Multi-modal Multi-label Recognition with Transformer","code":"https://github.com/iCVTEAM/M3TR","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"Q2L-TResL(resolution 448)","metrics":{"mAP":"96.1"},"uses_additional_data":false,"paper_date":"2021-07-22","paper":"/paper/query2label-a-simple-transformer-way-to-multi","paper_url":"https://arxiv.org/abs/2107.10834v1","paper_title":"Query2Label: A Simple Transformer Way to Multi-Label Classification","code":"https://github.com/SlongLiu/query2labels","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"MSRN(pretrain from MS-COCO)","metrics":{"mAP":"96.0"},"uses_additional_data":false,"paper_date":"2021-06-22","paper":"/paper/multi-layered-semantic-representation-network","paper_url":"https://arxiv.org/abs/2106.11596v1","paper_title":"Multi-layered Semantic Representation Network for Multi-label Image Classification","code":"https://github.com/chehao2628/MSRN","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"TResNet-L (resolution 448, pretrain from MS-COCO)","metrics":{"mAP":"95.8"},"uses_additional_data":false,"paper_date":"2020-09-29","paper":"/paper/asymmetric-loss-for-multi-label","paper_url":"https://arxiv.org/abs/2009.14119v4","paper_title":"Asymmetric Loss For Multi-Label Classification","code":"https://github.com/Alibaba-MIIL/ASL","n_code_links":5,"syntology":{"n_ran":4,"n_unverified":8,"n_samples":12,"n_pointer_only_licence":7}},{"rank_in_archive_order":9,"model":"SSGRL (pretrain from MS-COCO)","metrics":{"mAP":"95.0"},"uses_additional_data":false,"paper_date":"2019-08-20","paper":"/paper/learning-semantic-specific-graph","paper_url":"https://arxiv.org/abs/1908.07325v1","paper_title":"Learning Semantic-Specific Graph Representation for Multi-Label Image Recognition","code":"https://github.com/HCPLab-SYSU/SSGRL","n_code_links":2,"syntology":null},{"rank_in_archive_order":10,"model":"TDRG-R101(448×448)","metrics":{"mAP":"95.0"},"uses_additional_data":false,"paper_date":"2021-10-10","paper":"/paper/transformer-based-dual-relation-graph-for-1","paper_url":"https://arxiv.org/abs/2110.04722v2","paper_title":"Transformer-based Dual Relation Graph for Multi-label Image Recognition","code":"https://github.com/iCVTEAM/TDRG","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":3,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"MCAR (ResNet101, 448x448)","metrics":{"mAP":"94.8"},"uses_additional_data":false,"paper_date":"2020-07-03","paper":"/paper/multi-label-image-recognition-with-multi","paper_url":"https://arxiv.org/abs/2007.01755v3","paper_title":"Learning to Discover Multi-Class Attentional Regions for Multi-Label Image Recognition","code":"https://github.com/gaobb/MCAR","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"TResNet-L (resolution 448, pretrain from ImageNet)","metrics":{"mAP":"94.6"},"uses_additional_data":false,"paper_date":"2020-09-29","paper":"/paper/asymmetric-loss-for-multi-label","paper_url":"https://arxiv.org/abs/2009.14119v4","paper_title":"Asymmetric Loss For Multi-Label Classification","code":"https://github.com/Alibaba-MIIL/ASL","n_code_links":5,"syntology":{"n_ran":4,"n_unverified":8,"n_samples":12,"n_pointer_only_licence":7}},{"rank_in_archive_order":13,"model":"ML-GCN (pretrain from ImageNet)","metrics":{"mAP":"94.0"},"uses_additional_data":false,"paper_date":"2019-04-07","paper":"/paper/multi-label-image-recognition-with-graph","paper_url":"http://arxiv.org/abs/1904.03582v1","paper_title":"Multi-Label Image Recognition with Graph Convolutional Networks","code":"https://github.com/megvii-research/ml-gcn","n_code_links":2,"syntology":null},{"rank_in_archive_order":14,"model":"SSGRL (pretrain from ImageNet)","metrics":{"mAP":"93.4"},"uses_additional_data":false,"paper_date":"2019-08-20","paper":"/paper/learning-semantic-specific-graph","paper_url":"https://arxiv.org/abs/1908.07325v1","paper_title":"Learning Semantic-Specific Graph Representation for Multi-Label Image Recognition","code":"https://github.com/HCPLab-SYSU/SSGRL","n_code_links":2,"syntology":null},{"rank_in_archive_order":15,"model":"Ours PF-DLDL","metrics":{"mAP":"93.4"},"uses_additional_data":false,"paper_date":"2016-11-06","paper":"/paper/deep-label-distribution-learning-with-label","paper_url":"http://arxiv.org/abs/1611.01731v2","paper_title":"Deep Label Distribution Learning with Label Ambiguity","code":"https://github.com/gaobb/DLDL","n_code_links":2,"syntology":null},{"rank_in_archive_order":16,"model":"ViT-B-16 (ImageNet-21K pretrained)","metrics":{"mAP":"93.1"},"uses_additional_data":false,"paper_date":"2021-04-22","paper":"/paper/imagenet-21k-pretraining-for-the-masses","paper_url":"https://arxiv.org/abs/2104.10972v4","paper_title":"ImageNet-21K Pretraining for the Masses","code":"https://github.com/Alibaba-MIIL/ImageNet21K","n_code_links":5,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"FeV+LV  (pretrain from ImageNet)","metrics":{"mAP":"92.0"},"uses_additional_data":false,"paper_date":"2015-04-22","paper":"/paper/exploit-bounding-box-annotations-for-multi","paper_url":"http://arxiv.org/abs/1504.05843v2","paper_title":"Exploit Bounding Box Annotations for Multi-label Object Recognition","code":null,"n_code_links":0,"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 6,795 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":6795,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2785},"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":7,"rows_with_any_sample_ran":7,"distinct_papers_with_graph_line":4,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":16,"n_unverified":12,"n_samples":28,"n_pointer_only_licence":7,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":24,"n_unverified":20,"n_samples":44,"n_pointer_only_licence":14,"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"}}}