{"url":"/sota/unsupervised-few-shot-image-classification-on-1","task":{"name":"Unsupervised Few-Shot Image Classification","url":"/task/unsupervised-few-shot-image-classification","note":null},"dataset":{"name":"Mini-Imagenet 5-way (5-shot)","url":"/dataset/mini-imagenet"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"In contrast to (supervised) few-shot image classification, only the unlabeled dataset is available in the pre-training or meta-training stage for unsupervised few-shot image classification.","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":["Accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Accuracy":"higher"}},"counts":{"rows":28,"rows_with_code":16,"rows_with_paper_page":28,"rows_dated":28,"rows_using_additional_data":1},"rows":[{"rank_in_archive_order":1,"model":"BECLR","metrics":{"Accuracy":"87.82"},"uses_additional_data":false,"paper_date":"2024-02-04","paper":"/paper/beclr-batch-enhanced-contrastive-few-shot","paper_url":"https://arxiv.org/abs/2402.02444v1","paper_title":"BECLR: Batch Enhanced Contrastive Few-Shot Learning","code":"https://github.com/stypoumic/beclr","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"Meta-DM+UniSiam","metrics":{"Accuracy":"85.29"},"uses_additional_data":false,"paper_date":"2023-05-14","paper":"/paper/meta-dm-applications-of-diffusion-models-on","paper_url":"https://arxiv.org/abs/2305.08092v1","paper_title":"Meta-DM: Applications of Diffusion Models on Few-Shot Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"UniSiam","metrics":{"Accuracy":"83.40"},"uses_additional_data":false,"paper_date":"2022-07-19","paper":"/paper/self-supervision-can-be-a-good-few-shot","paper_url":"https://arxiv.org/abs/2207.09176v1","paper_title":"Self-Supervision Can Be a Good Few-Shot Learner","code":"https://github.com/bbbdylan/unisiam","n_code_links":4,"syntology":{"n_ran":5,"n_unverified":5,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"PDA-Net","metrics":{"Accuracy":"83.11"},"uses_additional_data":false,"paper_date":"2021-05-25","paper":"/paper/few-shot-learning-with-part-discovery-and","paper_url":"https://arxiv.org/abs/2105.11874v1","paper_title":"Few-Shot Learning with Part Discovery and Augmentation from Unlabeled Images","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"Deep Laplacian Eigenmaps","metrics":{"Accuracy":"78.79"},"uses_additional_data":false,"paper_date":"2022-10-07","paper":"/paper/unsupervised-few-shot-learning-via-deep","paper_url":"https://arxiv.org/abs/2210.03595v1","paper_title":"Unsupervised Few-shot Learning via Deep Laplacian Eigenmaps","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"UBC-FSL","metrics":{"Accuracy":"77.2"},"uses_additional_data":false,"paper_date":"2020-10-06","paper":"/paper/shot-in-the-dark-few-shot-learning-with-no-1","paper_url":"https://arxiv.org/abs/2010.02430v2","paper_title":"Shot in the Dark: Few-Shot Learning with No Base-Class Labels","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"HMS","metrics":{"Accuracy":"75.77"},"uses_additional_data":false,"paper_date":"2020-11-30","paper":"/paper/revisiting-unsupervised-meta-learning","paper_url":"https://arxiv.org/abs/2011.14663v3","paper_title":"Revisiting Unsupervised Meta-Learning via the Characteristics of Few-Shot Tasks","code":"https://github.com/hanlu-nju/revisiting-uml","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"TrainProto","metrics":{"Accuracy":"73.94"},"uses_additional_data":false,"paper_date":"2021-06-21","paper":"/paper/trainable-class-prototypes-for-few-shot","paper_url":"https://arxiv.org/abs/2106.10846v1","paper_title":"Trainable Class Prototypes for Few-Shot Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"CPNWCP","metrics":{"Accuracy":"73.21"},"uses_additional_data":false,"paper_date":"2022-10-21","paper":"/paper/contrastive-prototypical-network-with","paper_url":"https://www.ecva.net/papers/eccv_2022/papers_ECCV/html/121_ECCV_2022_paper.php","paper_title":"Contrastive Prototypical Network with Wasserstein Confidence Penalty","code":"https://github.com/Haoqing-Wang/CPNWCP","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"SAMPTransfer (Conv4)","metrics":{"Accuracy":"72.52"},"uses_additional_data":false,"paper_date":"2022-10-12","paper":"/paper/self-attention-message-passing-for","paper_url":"https://arxiv.org/abs/2210.06339v1","paper_title":"Self-Attention Message Passing for Contrastive Few-Shot Learning","code":"https://github.com/ojss/samptransfer","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"AmdimNet","metrics":{"Accuracy":"70.14"},"uses_additional_data":false,"paper_date":"2019-11-14","paper":"/paper/self-supervised-learning-for-few-shot-image","paper_url":"https://arxiv.org/abs/1911.06045v3","paper_title":"Self-Supervised Learning For Few-Shot Image Classification","code":"https://github.com/Alibaba-AAIG/SSL-FEW-SHOT","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"CSSL","metrics":{"Accuracy":"68.91"},"uses_additional_data":true,"paper_date":"2020-08-23","paper":"/paper/few-shot-image-classification-via-contrastive","paper_url":"https://arxiv.org/abs/2008.09942v1","paper_title":"Few-Shot Image Classification via Contrastive Self-Supervised Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"LF2CS","metrics":{"Accuracy":"67.36"},"uses_additional_data":false,"paper_date":"2022-10-21","paper":"/paper/unsupervised-few-shot-image-classification-by","paper_url":"https://www.ecva.net/papers/eccv_2022/papers_ECCV/html/7167_ECCV_2022_paper.php","paper_title":"Unsupervised Few-Shot Image Classification by Learning Features into Clustering Space","code":"https://github.com/xidianai/LF2CS","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"C^3LR","metrics":{"Accuracy":"64.81"},"uses_additional_data":false,"paper_date":"2022-02-15","paper":"/paper/self-supervised-class-cognizant-few-shot","paper_url":"https://arxiv.org/abs/2202.08149v1","paper_title":"Self-Supervised Class-Cognizant Few-Shot Classification","code":"https://github.com/ojss/c3lr","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"PsCo","metrics":{"Accuracy":"63.26"},"uses_additional_data":false,"paper_date":"2023-03-02","paper":"/paper/unsupervised-meta-learning-via-few-shot-1","paper_url":"https://arxiv.org/abs/2303.00996v1","paper_title":"Unsupervised Meta-Learning via Few-shot Pseudo-supervised Contrastive Learning","code":"https://github.com/alinlab/psco","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"ProtoTransfer","metrics":{"Accuracy":"62.99"},"uses_additional_data":false,"paper_date":"2020-06-19","paper":"/paper/self-supervised-prototypical-transfer","paper_url":"https://arxiv.org/abs/2006.11325v1","paper_title":"Self-Supervised Prototypical Transfer Learning for Few-Shot Classification","code":"https://github.com/indy-lab/ProtoTransfer","n_code_links":2,"syntology":{"n_ran":8,"n_unverified":1,"n_samples":9,"n_pointer_only_licence":4}},{"rank_in_archive_order":17,"model":"PL-CFE","metrics":{"Accuracy":"62.91"},"uses_additional_data":false,"paper_date":"2022-09-27","paper":"/paper/rethinking-clustering-based-pseudo-labeling","paper_url":"https://arxiv.org/abs/2209.13635v1","paper_title":"Rethinking Clustering-Based Pseudo-Labeling for Unsupervised Meta-Learning","code":"https://github.com/xingpingdong/pl-cfe","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"CMVAE","metrics":{"Accuracy":"58.95"},"uses_additional_data":false,"paper_date":"2023-02-20","paper":"/paper/cmvae-causal-meta-vae-for-unsupervised-meta","paper_url":"https://arxiv.org/abs/2302.09731v1","paper_title":"CMVAE: Causal Meta VAE for Unsupervised Meta-Learning","code":"https://github.com/guodongqi/cmvae","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":19,"model":"Meta-SVEBM","metrics":{"Accuracy":"58.03"},"uses_additional_data":false,"paper_date":"2021-09-30","paper":"/paper/unsupervised-meta-learning-via-latent-space","paper_url":"https://openreview.net/forum?id=-pLftu7EpXz","paper_title":"Unsupervised Meta-Learning via Latent Space Energy-based Model of Symbol Vector Coupling","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":20,"model":"ArL","metrics":{"Accuracy":"57.01"},"uses_additional_data":false,"paper_date":"2020-01-12","paper":"/paper/rethinking-class-relations-absolute-relative","paper_url":"https://arxiv.org/abs/2001.03919v4","paper_title":"Rethinking Class Relations: Absolute-relative Supervised and Unsupervised Few-shot Learning","code":"https://github.com/ojss/samptransfer","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"InCo","metrics":{"Accuracy":"56.96"},"uses_additional_data":false,"paper_date":"2022-11-01","paper":"/paper/invariant-and-consistent-unsupervised","paper_url":"https://www.sciencedirect.com/science/article/pii/S0925231222014692","paper_title":"Invariant and consistent: Unsupervised representation learning for few-shot visual recognition","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":22,"model":"ULDA","metrics":{"Accuracy":"56.18"},"uses_additional_data":false,"paper_date":"2020-04-13","paper":"/paper/unsupervised-few-shot-learning-via","paper_url":"https://arxiv.org/abs/2004.05805v2","paper_title":"Diversity Helps: Unsupervised Few-shot Learning via Distribution Shift-based Data Augmentation","code":"https://github.com/WonderSeven/ULDA","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":5,"n_samples":13,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"Meta-GMVAE","metrics":{"Accuracy":"55.73"},"uses_additional_data":false,"paper_date":"2021-01-01","paper":"/paper/meta-gmvae-mixture-of-gaussian-vae-for","paper_url":"https://openreview.net/forum?id=wS0UFjsNYjn","paper_title":"Meta-GMVAE: Mixture of Gaussian VAE for Unsupervised Meta-Learning","code":"https://github.com/db-Lee/Meta-GMVAE","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"U-MlSo+PN","metrics":{"Accuracy":"55.38"},"uses_additional_data":false,"paper_date":"2022-01-15","paper":"/paper/multi-level-second-order-few-shot-learning","paper_url":"https://arxiv.org/abs/2201.05916v1","paper_title":"Multi-level Second-order Few-shot Learning","code":"https://github.com/hongguangzhang/mlso-tmm-master","n_code_links":1,"syntology":null},{"rank_in_archive_order":25,"model":"LASIUM","metrics":{"Accuracy":"54.56"},"uses_additional_data":false,"paper_date":"2020-06-18","paper":"/paper/unsupervised-meta-learning-through-latent","paper_url":"https://arxiv.org/abs/2006.10236v1","paper_title":"Unsupervised Meta-Learning through Latent-Space Interpolation in Generative Models","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":26,"model":"CACTU","metrics":{"Accuracy":"53.97"},"uses_additional_data":false,"paper_date":"2018-10-04","paper":"/paper/unsupervised-learning-via-meta-learning","paper_url":"http://arxiv.org/abs/1810.02334v6","paper_title":"Unsupervised Learning via Meta-Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":27,"model":"UMTRA","metrics":{"Accuracy":"50.73"},"uses_additional_data":false,"paper_date":"2018-11-28","paper":"/paper/unsupervised-meta-learning-for-few-shot-image","paper_url":"https://arxiv.org/abs/1811.11819v2","paper_title":"Unsupervised Meta-Learning For Few-Shot Image Classification","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":28,"model":"AAL","metrics":{"Accuracy":"49.18"},"uses_additional_data":false,"paper_date":"2019-02-26","paper":"/paper/assume-augment-and-learn-unsupervised-few","paper_url":"http://arxiv.org/abs/1902.09884v3","paper_title":"Assume, Augment and Learn: Unsupervised Few-Shot Meta-Learning via Random Labels and Data Augmentation","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 6,821 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":6821,"papers_extracted_not_yet_verified":65,"boards_without_verdict":29,"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":7,"rows_with_any_sample_ran":6,"distinct_papers_with_graph_line":7,"distinct_papers_with_any_sample_ran":6,"samples_over_distinct_papers":{"n_ran":27,"n_unverified":12,"n_samples":39,"n_pointer_only_licence":6,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":27,"n_unverified":12,"n_samples":39,"n_pointer_only_licence":6,"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"}}}