{"url":"/sota/semi-supervised-image-classification-on-cifar-8","task":{"name":"Semi-Supervised Image Classification","url":"/task/semi-supervised-image-classification","note":null},"dataset":{"name":"CIFAR-100, 400 Labels","url":"/dataset/cifar-100"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"Semi-supervised image classification leverages unlabelled data as well as labelled data to increase classification performance.\r\n\r\nYou may want to read some blog posts to get an overview before reading the papers and checking the leaderboards:\r\n\r\n- [An overview of proxy-label approaches for semi-supervised learning](https://ruder.io/semi-supervised/) - Sebastian Ruder\r\n- [Semi-Supervised Learning in Computer Vision](https://amitness.com/2020/07/semi-supervised-learning/) - Amit Chaudhary\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Self-Supervised Semi-Supervised Learning](https://arxiv.org/pdf/1905.03670v2.pdf) )</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":["Percentage error"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Percentage error":"lower"}},"counts":{"rows":21,"rows_with_code":14,"rows_with_paper_page":20,"rows_dated":20,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"SemiReward","metrics":{"Percentage error":"15.62"},"uses_additional_data":false,"paper_date":"2023-10-04","paper":"/paper/semireward-a-general-reward-model-for-semi","paper_url":"https://arxiv.org/abs/2310.03013v2","paper_title":"SemiReward: A General Reward Model for Semi-supervised Learning","code":"https://github.com/Westlake-AI/SemiReward","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":1,"n_samples":5,"n_pointer_only_licence":1}},{"rank_in_archive_order":2,"model":"ReMixMatch","metrics":{"Percentage error":"16.8"},"uses_additional_data":false,"paper_date":"2022-08-12","paper":"/paper/usb-a-unified-semi-supervised-learning","paper_url":"https://arxiv.org/abs/2208.07204v2","paper_title":"USB: A Unified Semi-supervised Learning Benchmark for Classification","code":"https://github.com/microsoft/semi-supervised-learning","n_code_links":5,"syntology":{"n_ran":9,"n_unverified":0,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"SemiOccam","metrics":{"Percentage error":"26.59"},"uses_additional_data":false,"paper_date":"2025-06-04","paper":"/paper/vitsgmm-a-robust-semi-supervised-image-1","paper_url":"https://arxiv.org/abs/2506.03582v1","paper_title":"ViTSGMM: A Robust Semi-Supervised Image Recognition Network Using Sparse Labels","code":"https://github.com/Shu1L0n9/SemiOccam","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"Semi-SST (ViT-Small)","metrics":{"Percentage error":"31.39±0.47"},"uses_additional_data":false,"paper_date":"2025-05-31","paper":"/paper/sst-self-training-with-self-adaptive-1","paper_url":"https://arxiv.org/abs/2506.00467v1","paper_title":"SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"ShrinkMatch","metrics":{"Percentage error":"35.36"},"uses_additional_data":false,"paper_date":"2023-08-13","paper":"/paper/shrinking-class-space-for-enhanced-certainty","paper_url":"https://arxiv.org/abs/2308.06777v1","paper_title":"Shrinking Class Space for Enhanced Certainty in Semi-Supervised Learning","code":"https://github.com/LiheYoung/ShrinkMatch","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"Super-SST (ViT-Small)","metrics":{"Percentage error":"35.50±0.58"},"uses_additional_data":false,"paper_date":"2025-05-31","paper":"/paper/sst-self-training-with-self-adaptive-1","paper_url":"https://arxiv.org/abs/2506.00467v1","paper_title":"SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"PCL (Flexmatch)","metrics":{"Percentage error":"35.75±0.53"},"uses_additional_data":false,"paper_date":"2021-11-11","paper":"/paper/semantic-aware-representation-learning-via-1","paper_url":"https://arxiv.org/abs/2111.06021v5","paper_title":"Probabilistic Contrastive Learning for Domain Adaptation","code":"https://github.com/ljjcoder/PCL","n_code_links":2,"syntology":null},{"rank_in_archive_order":8,"model":"SimMatch","metrics":{"Percentage error":"37.81"},"uses_additional_data":false,"paper_date":"2022-03-14","paper":"/paper/simmatch-semi-supervised-learning-with","paper_url":"https://arxiv.org/abs/2203.06915v2","paper_title":"SimMatch: Semi-supervised Learning with Similarity Matching","code":"https://github.com/kylezheng1997/simmatch","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"FreeMatch","metrics":{"Percentage error":"37.98"},"uses_additional_data":false,"paper_date":"2022-05-15","paper":"/paper/freematch-self-adaptive-thresholding-for-semi","paper_url":"https://arxiv.org/abs/2205.07246v3","paper_title":"FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning","code":"https://github.com/microsoft/semi-supervised-learning","n_code_links":6,"syntology":{"n_ran":9,"n_unverified":5,"n_samples":14,"n_pointer_only_licence":2}},{"rank_in_archive_order":10,"model":"NP-Match","metrics":{"Percentage error":"38.67"},"uses_additional_data":false,"paper_date":"2022-07-03","paper":"/paper/np-match-when-neural-processes-meet-semi","paper_url":"https://arxiv.org/abs/2207.01066v1","paper_title":"NP-Match: When Neural Processes meet Semi-Supervised Learning","code":"https://github.com/jianf-wang/np-match","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":11,"model":"CCSSL(FixMatch)","metrics":{"Percentage error":"38.81"},"uses_additional_data":false,"paper_date":"2022-03-04","paper":"/paper/class-aware-contrastive-semi-supervised","paper_url":"https://arxiv.org/abs/2203.02261v3","paper_title":"Class-Aware Contrastive Semi-Supervised Learning","code":"https://github.com/tencentyouturesearch/classification-semicls","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":12,"model":"FlexMatch","metrics":{"Percentage error":"39.94±1.62"},"uses_additional_data":false,"paper_date":"2021-10-15","paper":"/paper/flexmatch-boosting-semi-supervised-learning","paper_url":"https://arxiv.org/abs/2110.08263v3","paper_title":"FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling","code":"https://github.com/torchssl/torchssl","n_code_links":2,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"FixMatch+DM","metrics":{"Percentage error":"40.25±0.95"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":14,"model":"DoubleMatch","metrics":{"Percentage error":"41.83± 1.22"},"uses_additional_data":false,"paper_date":"2022-05-11","paper":"/paper/doublematch-improving-semi-supervised","paper_url":"https://arxiv.org/abs/2205.05575v1","paper_title":"DoubleMatch: Improving Semi-Supervised Learning with Self-Supervision","code":"https://github.com/walline/doublematch","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"PCL (Fixmatch)","metrics":{"Percentage error":"42.38±2.52"},"uses_additional_data":false,"paper_date":"2021-11-11","paper":"/paper/semantic-aware-representation-learning-via-1","paper_url":"https://arxiv.org/abs/2111.06021v5","paper_title":"Probabilistic Contrastive Learning for Domain Adaptation","code":"https://github.com/ljjcoder/PCL","n_code_links":2,"syntology":null},{"rank_in_archive_order":16,"model":"DP-SSL","metrics":{"Percentage error":"43.17±1.29"},"uses_additional_data":false,"paper_date":"2021-10-26","paper":"/paper/dp-ssl-towards-robust-semi-supervised","paper_url":"https://arxiv.org/abs/2110.13740v1","paper_title":"DP-SSL: Towards Robust Semi-supervised Learning with A Few Labeled Samples","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":17,"model":"ReMixMatch","metrics":{"Percentage error":"44.28±2.06"},"uses_additional_data":false,"paper_date":"2019-11-21","paper":"/paper/remixmatch-semi-supervised-learning-with-1","paper_url":"https://arxiv.org/abs/1911.09785v2","paper_title":"ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring","code":"https://github.com/google-research/mixmatch","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":3,"n_samples":4,"n_pointer_only_licence":1}},{"rank_in_archive_order":18,"model":"Dash\n(RA, WRN-28-8)","metrics":{"Percentage error":"44.76±0.96"},"uses_additional_data":false,"paper_date":"2021-09-01","paper":"/paper/dash-semi-supervised-learning-with-dynamic","paper_url":"https://arxiv.org/abs/2109.00650v1","paper_title":"Dash: Semi-Supervised Learning with Dynamic Thresholding","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":19,"model":"Dash\n(CTA, WRN-28-8)","metrics":{"Percentage error":"44.83±1.36"},"uses_additional_data":false,"paper_date":"2021-09-01","paper":"/paper/dash-semi-supervised-learning-with-dynamic","paper_url":"https://arxiv.org/abs/2109.00650v1","paper_title":"Dash: Semi-Supervised Learning with Dynamic Thresholding","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":20,"model":"FixMatch+CR","metrics":{"Percentage error":"49.23"},"uses_additional_data":false,"paper_date":"2022-01-17","paper":"/paper/contrastive-regularization-for-semi","paper_url":"https://arxiv.org/abs/2201.06247v2","paper_title":"Contrastive Regularization for Semi-Supervised Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":21,"model":"FixMatch (CTA)","metrics":{"Percentage error":"49.95±3.01"},"uses_additional_data":false,"paper_date":"2020-01-21","paper":"/paper/fixmatch-simplifying-semi-supervised-learning","paper_url":"https://arxiv.org/abs/2001.07685v2","paper_title":"FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence","code":"https://github.com/google-research/fixmatch","n_code_links":26,"syntology":{"n_ran":52,"n_unverified":22,"n_samples":74,"n_pointer_only_licence":14}}],"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 6,885 of the 9,623 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":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+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":9,"rows_with_any_sample_ran":9,"distinct_papers_with_graph_line":9,"distinct_papers_with_any_sample_ran":9,"samples_over_distinct_papers":{"n_ran":83,"n_unverified":35,"n_samples":118,"n_pointer_only_licence":22,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":83,"n_unverified":35,"n_samples":118,"n_pointer_only_licence":22,"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"}}}