{"url":"/sota/weakly-supervised-action-localization-on-4","task":{"name":"Weakly Supervised Action Localization","url":"/task/weakly-supervised-action-localization","note":null},"dataset":{"name":"THUMOS’14","url":"/dataset/thumos14-1"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"In this task, the training data consists of videos with a list of activities in them without any temporal boundary annotations. However, while testing, given a video, the algorithm should recognize the activities in the video and also provide the start and end time.","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@0.5","avg-mAP (0.1-0.5)"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"mAP@0.5":"higher","avg-mAP (0.1-0.5)":"higher"}},"counts":{"rows":13,"rows_with_code":11,"rows_with_paper_page":13,"rows_dated":13,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"LACP","metrics":{"mAP@0.5":"45.3"},"uses_additional_data":false,"paper_date":"2021-08-11","paper":"/paper/learning-action-completeness-from-points-for","paper_url":"https://arxiv.org/abs/2108.05029v1","paper_title":"Learning Action Completeness from Points for Weakly-supervised Temporal Action Localization","code":"https://github.com/Pilhyeon/Learning-Action-Completeness-from-Points","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"ProCL","metrics":{"avg-mAP (0.1-0.5)":"58.2","mAP@0.5":"40.5"},"uses_additional_data":false,"paper_date":"2022-06-22","paper":"/paper/weakly-supervised-action-localization-via","paper_url":"https://arxiv.org/abs/2206.11011v2","paper_title":"Weakly-Supervised Temporal Action Localization by Progressive Complementary Learning","code":"https://github.com/Run542968/ProCL","n_code_links":2,"syntology":null},{"rank_in_archive_order":3,"model":"P-MIL","metrics":{"mAP@0.5":"40.0"},"uses_additional_data":false,"paper_date":"2023-05-29","paper":"/paper/proposal-based-multiple-instance-learning-for-1","paper_url":"https://arxiv.org/abs/2305.17861v1","paper_title":"Proposal-Based Multiple Instance Learning for Weakly-Supervised Temporal Action Localization","code":"https://github.com/RenHuan1999/CVPR2023_P-MIL","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"D2-Net","metrics":{"mAP@0.5":"35.9"},"uses_additional_data":false,"paper_date":"2020-12-11","paper":"/paper/d2-net-weakly-supervised-action-localization","paper_url":"https://arxiv.org/abs/2012.06440v2","paper_title":"D2-Net: Weakly-Supervised Action Localization via Discriminative Embeddings and Denoised Activations","code":"https://github.com/naraysa/D2-Net","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"Ju et al.","metrics":{"mAP@0.5":"35.9"},"uses_additional_data":false,"paper_date":"2020-12-15","paper":"/paper/point-level-temporal-action-localization","paper_url":"https://arxiv.org/abs/2012.08236v1","paper_title":"Point-Level Temporal Action Localization: Bridging Fully-supervised Proposals to Weakly-supervised Losses","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"ACM-Net","metrics":{"mAP@0.5":"34.6"},"uses_additional_data":false,"paper_date":"2021-04-07","paper":"/paper/acm-net-action-context-modeling-network-for","paper_url":"https://arxiv.org/abs/2104.02967v1","paper_title":"ACM-Net: Action Context Modeling Network for Weakly-Supervised Temporal Action Localization","code":"https://github.com/ispc-lab/ACM-Net","n_code_links":2,"syntology":null},{"rank_in_archive_order":7,"model":"Lee et al.","metrics":{"mAP@0.5":"33.7"},"uses_additional_data":false,"paper_date":"2020-06-12","paper":"/paper/background-modeling-via-uncertainty","paper_url":"https://arxiv.org/abs/2006.07006v3","paper_title":"Weakly-supervised Temporal Action Localization by Uncertainty Modeling","code":"https://github.com/Pilhyeon/WTAL-Uncertainty-Modeling","n_code_links":2,"syntology":null},{"rank_in_archive_order":8,"model":"ACS-Net","metrics":{"mAP@0.5":"32.4"},"uses_additional_data":false,"paper_date":"2021-03-28","paper":"/paper/acsnet-action-context-separation-network-for","paper_url":"https://arxiv.org/abs/2103.15088v1","paper_title":"ACSNet: Action-Context Separation Network for Weakly Supervised Temporal Action Localization","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"CoLA","metrics":{"mAP@0.5":"32.2"},"uses_additional_data":false,"paper_date":"2021-03-30","paper":"/paper/cola-weakly-supervised-temporal-action","paper_url":"https://arxiv.org/abs/2103.16392v2","paper_title":"CoLA: Weakly-Supervised Temporal Action Localization with Snippet Contrastive Learning","code":"https://github.com/zhang-can/CoLA","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"EM-MIL","metrics":{"mAP@0.5":"30.5"},"uses_additional_data":false,"paper_date":"2020-03-31","paper":"/paper/weakly-supervised-action-localization-with-2","paper_url":"https://arxiv.org/abs/2004.00163v2","paper_title":"Weakly-Supervised Action Localization with Expectation-Maximization Multi-Instance Learning","code":"https://github.com/airmachine/EM-MIL-WeaklyActionDetection","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"A2CL-PT","metrics":{"mAP@0.5":"30.1"},"uses_additional_data":false,"paper_date":"2020-07-13","paper":"/paper/adversarial-background-aware-loss-for-weakly","paper_url":"https://arxiv.org/abs/2007.06643v1","paper_title":"Adversarial Background-Aware Loss for Weakly-supervised Temporal Activity Localization","code":"https://github.com/kylemin/A2CL-PT","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"BasNet","metrics":{"mAP@0.5":"27.0"},"uses_additional_data":false,"paper_date":"2019-11-22","paper":"/paper/background-suppression-network-for-weakly","paper_url":"https://arxiv.org/abs/1911.09963v1","paper_title":"Background Suppression Network for Weakly-supervised Temporal Action Localization","code":"https://github.com/Pilhyeon/BaSNet-pytorch","n_code_links":2,"syntology":null},{"rank_in_archive_order":13,"model":"3C-Net","metrics":{"mAP@0.5":"26.6"},"uses_additional_data":false,"paper_date":"2019-08-22","paper":"/paper/3c-net-category-count-and-center-loss-for","paper_url":"https://arxiv.org/abs/1908.08216v2","paper_title":"3C-Net: Category Count and Center Loss for Weakly-Supervised Action Localization","code":"https://github.com/naraysa/3c-net","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":0,"rows_with_any_sample_ran":0,"distinct_papers_with_graph_line":0,"distinct_papers_with_any_sample_ran":0,"samples_over_distinct_papers":{"n_ran":0,"n_unverified":0,"n_samples":0,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":0,"n_unverified":0,"n_samples":0,"n_pointer_only_licence":0,"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"}}}