{"url":"/sota/few-shot-image-classification-on-meta-dataset-1","task":{"name":"Few-Shot Image Classification","url":"/task/few-shot-image-classification","note":null},"dataset":{"name":"Meta-Dataset Rank","url":"/dataset/meta-dataset"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"**Few-Shot Image Classification** is a computer vision task that involves training machine learning models to classify images into predefined categories using only a few labeled examples of each category (typically < 6 examples). The goal is to enable models to recognize and classify new images with minimal supervision and limited data, without having to train on large datasets. (typically < 6 examples)\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Learning Embedding Adaptation for Few-Shot Learning](https://github.com/Sha-Lab/FEAT) )</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":["Mean Rank"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Mean Rank":null}},"counts":{"rows":13,"rows_with_code":13,"rows_with_paper_page":13,"rows_dated":13,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"URT","metrics":{"Mean Rank":"2.85"},"uses_additional_data":false,"paper_date":"2020-06-21","paper":"/paper/a-universal-representation-transformer-layer","paper_url":"https://arxiv.org/abs/2006.11702v4","paper_title":"A Universal Representation Transformer Layer for Few-Shot Image Classification","code":"https://github.com/liulu112601/URT","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"Transductive CNAPS","metrics":{"Mean Rank":"3.05"},"uses_additional_data":false,"paper_date":"2020-06-17","paper":"/paper/improving-few-shot-visual-classification-with","paper_url":"https://arxiv.org/abs/2006.12245v6","paper_title":"Enhancing Few-Shot Image Classification with Unlabelled Examples","code":"https://github.com/peymanbateni/simple-cnaps","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":6,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"Simple CNAPS","metrics":{"Mean Rank":"3.45"},"uses_additional_data":false,"paper_date":"2019-12-07","paper":"/paper/improved-few-shot-visual-classification","paper_url":"https://arxiv.org/abs/1912.03432v3","paper_title":"Improved Few-Shot Visual Classification","code":"https://github.com/peymanbateni/simple-cnaps","n_code_links":2,"syntology":null},{"rank_in_archive_order":4,"model":"SUR","metrics":{"Mean Rank":"4.2"},"uses_additional_data":false,"paper_date":"2020-03-20","paper":"/paper/selecting-relevant-features-from-a-universal","paper_url":"https://arxiv.org/abs/2003.09338v2","paper_title":"Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification","code":"https://github.com/dvornikita/SUR","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":2,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"SUR-pnf","metrics":{"Mean Rank":"4.25"},"uses_additional_data":false,"paper_date":"2020-03-20","paper":"/paper/selecting-relevant-features-from-a-universal","paper_url":"https://arxiv.org/abs/2003.09338v2","paper_title":"Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification","code":"https://github.com/dvornikita/SUR","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":2,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"CNAPs","metrics":{"Mean Rank":"5.95"},"uses_additional_data":false,"paper_date":"2019-06-18","paper":"/paper/fast-and-flexible-multi-task-classification","paper_url":"https://arxiv.org/abs/1906.07697v2","paper_title":"Fast and Flexible Multi-Task Classification Using Conditional Neural Adaptive Processes","code":"https://github.com/cambridge-mlg/cnaps","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"fo-Proto-MAML","metrics":{"Mean Rank":"6.65"},"uses_additional_data":false,"paper_date":"2019-03-07","paper":"/paper/meta-dataset-a-dataset-of-datasets-for","paper_url":"https://arxiv.org/abs/1903.03096v4","paper_title":"Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples","code":"https://github.com/google-research/meta-dataset","n_code_links":15,"syntology":{"n_ran":10,"n_unverified":1,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"Prototypical Networks","metrics":{"Mean Rank":"8.5"},"uses_additional_data":false,"paper_date":"2017-03-15","paper":"/paper/prototypical-networks-for-few-shot-learning","paper_url":"http://arxiv.org/abs/1703.05175v2","paper_title":"Prototypical Networks for Few-shot Learning","code":"https://github.com/learnables/learn2learn","n_code_links":43,"syntology":{"n_ran":49,"n_unverified":15,"n_samples":64,"n_pointer_only_licence":18}},{"rank_in_archive_order":9,"model":"Finetune","metrics":{"Mean Rank":"8.7"},"uses_additional_data":false,"paper_date":"2019-03-07","paper":"/paper/meta-dataset-a-dataset-of-datasets-for","paper_url":"https://arxiv.org/abs/1903.03096v4","paper_title":"Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples","code":"https://github.com/google-research/meta-dataset","n_code_links":15,"syntology":{"n_ran":10,"n_unverified":1,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"fo-MAML","metrics":{"Mean Rank":"10.25"},"uses_additional_data":false,"paper_date":"2017-03-09","paper":"/paper/model-agnostic-meta-learning-for-fast","paper_url":"http://arxiv.org/abs/1703.03400v3","paper_title":"Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks","code":"https://github.com/ray-project/ray/tree/master/rllib","n_code_links":85,"syntology":{"n_ran":90,"n_unverified":64,"n_samples":154,"n_pointer_only_licence":57}},{"rank_in_archive_order":11,"model":"Matching Networks","metrics":{"Mean Rank":"10.5"},"uses_additional_data":false,"paper_date":"2016-06-13","paper":"/paper/matching-networks-for-one-shot-learning","paper_url":"http://arxiv.org/abs/1606.04080v2","paper_title":"Matching Networks for One Shot Learning","code":"https://github.com/oscarknagg/few-shot","n_code_links":26,"syntology":{"n_ran":6,"n_unverified":10,"n_samples":16,"n_pointer_only_licence":6}},{"rank_in_archive_order":12,"model":"k-NN","metrics":{"Mean Rank":"10.85"},"uses_additional_data":false,"paper_date":"2019-03-07","paper":"/paper/meta-dataset-a-dataset-of-datasets-for","paper_url":"https://arxiv.org/abs/1903.03096v4","paper_title":"Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples","code":"https://github.com/google-research/meta-dataset","n_code_links":15,"syntology":{"n_ran":10,"n_unverified":1,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"Relation Networks","metrics":{"Mean Rank":"11.8"},"uses_additional_data":false,"paper_date":"2017-11-16","paper":"/paper/learning-to-compare-relation-network-for-few","paper_url":"http://arxiv.org/abs/1711.06025v2","paper_title":"Learning to Compare: Relation Network for Few-Shot Learning","code":"https://github.com/sicara/easy-few-shot-learning","n_code_links":13,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":1}}],"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":10,"rows_with_any_sample_ran":10,"distinct_papers_with_graph_line":7,"distinct_papers_with_any_sample_ran":7,"samples_over_distinct_papers":{"n_ran":167,"n_unverified":99,"n_samples":266,"n_pointer_only_licence":82,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":195,"n_unverified":103,"n_samples":298,"n_pointer_only_licence":82,"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"}}}