{"url":"/sota/image-classification-on-fashion-mnist","task":{"name":"Image Classification","url":"/task/image-classification","note":null},"dataset":{"name":"Fashion-MNIST","url":"/dataset/fashion-mnist"},"category":"Computer Vision","categories":["Adversarial","Computer Vision"],"category_note":null,"description":"**Image Classification** is a fundamental task in vision recognition that aims to understand and categorize an image as a whole under a specific label. Unlike [object detection](/task/object-detection), which involves classification and location of multiple objects within an image, image classification typically pertains to single-object images. When the classification becomes highly detailed or reaches instance-level, it is often referred to as [image retrieval](/task/image-retrieval), which also involves finding similar images in a large database.\r\n\r\n\r\n<span class=\"description-source\">Source: [Metamorphic Testing for Object Detection Systems ](https://arxiv.org/abs/1912.12162)</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","Accuracy","Trainable Parameters","NMI","Power consumption"],"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","Accuracy":"higher","Trainable Parameters":null,"NMI":null,"Power consumption":null}},"counts":{"rows":34,"rows_with_code":28,"rows_with_paper_page":33,"rows_dated":33,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"PreAct-ResNet18 + FMix","metrics":{"Percentage error":"3.64"},"uses_additional_data":false,"paper_date":"2020-02-27","paper":"/paper/understanding-and-enhancing-mixed-sample-data","paper_url":"https://arxiv.org/abs/2002.12047v3","paper_title":"FMix: Enhancing Mixed Sample Data Augmentation","code":"https://github.com/PaddlePaddle/PaddleClas","n_code_links":5,"syntology":{"n_ran":8,"n_unverified":0,"n_samples":8,"n_pointer_only_licence":3}},{"rank_in_archive_order":2,"model":"Random Erasing","metrics":{"Percentage error":"3.65"},"uses_additional_data":false,"paper_date":"2017-08-16","paper":"/paper/random-erasing-data-augmentation","paper_url":"http://arxiv.org/abs/1708.04896v2","paper_title":"Random Erasing Data Augmentation","code":"https://github.com/rwightman/pytorch-image-models","n_code_links":18,"syntology":null},{"rank_in_archive_order":3,"model":"E2E-3M","metrics":{"Percentage error":"4.08"},"uses_additional_data":false,"paper_date":"2020-07-30","paper":"/paper/rethinking-recurrent-neural-networks-and","paper_url":"https://arxiv.org/abs/2007.15161v3","paper_title":"Rethinking Recurrent Neural Networks and Other Improvements for Image Classification","code":"https://github.com/leonlha/e2e-3m","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"VGG8B(2x) + LocalLearning + CO","metrics":{"Percentage error":"4.14"},"uses_additional_data":false,"paper_date":"2019-01-20","paper":"/paper/training-neural-networks-with-local-error","paper_url":"https://arxiv.org/abs/1901.06656v2","paper_title":"Training Neural Networks with Local Error Signals","code":"https://github.com/anokland/local-loss","n_code_links":2,"syntology":null},{"rank_in_archive_order":5,"model":"Inception v3","metrics":{"Accuracy":"94.44","Percentage error":"5.56"},"uses_additional_data":false,"paper_date":"2022-03-29","paper":"/paper/cnn-filter-db-an-empirical-investigation-of","paper_url":"https://arxiv.org/abs/2203.15331v2","paper_title":"CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters","code":"https://github.com/paulgavrikov/cnn-filter-db","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"WaveMixLite","metrics":{"Percentage error":"5.68"},"uses_additional_data":false,"paper_date":"2022-05-28","paper":"/paper/wavemix-lite-a-resource-efficient-neural","paper_url":"https://arxiv.org/abs/2205.14375v5","paper_title":"WaveMix: A Resource-efficient Neural Network for Image Analysis","code":"https://github.com/pranavphoenix/WaveMix","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"Local Mixup DenseNet","metrics":{"Percentage error":"5.97"},"uses_additional_data":false,"paper_date":"2022-01-12","paper":"/paper/preventing-manifold-intrusion-with-locality","paper_url":"https://arxiv.org/abs/2201.04368v1","paper_title":"Preventing Manifold Intrusion with Locality: Local Mixup","code":"https://github.com/raphael-baena/Local-Mixup","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"TextCaps","metrics":{"Percentage error":"6.29"},"uses_additional_data":false,"paper_date":"2019-04-17","paper":"/paper/textcaps-handwritten-character-recognition","paper_url":"http://arxiv.org/abs/1904.08095v1","paper_title":"TextCaps : Handwritten Character Recognition with Very Small Datasets","code":"https://github.com/vinojjayasundara/textcaps","n_code_links":3,"syntology":null},{"rank_in_archive_order":9,"model":"R-ExplaiNet-26","metrics":{"Accuracy":"93.45","Percentage error":"6.55","Trainable Parameters":"892362"},"uses_additional_data":false,"paper_date":"2024-10-31","paper":"/paper/learning-local-discrete-features-in","paper_url":"https://arxiv.org/abs/2411.00139v1","paper_title":"Learning local discrete features in explainable-by-design convolutional neural networks","code":"https://github.com/pikaplan/LearnExplaiNet","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"ResNet-18 + Vision Eagle Attention","metrics":{"Accuracy":"93.30","Percentage error":"6.70"},"uses_additional_data":false,"paper_date":"2024-11-15","paper":"/paper/vision-eagle-attention-a-new-lens-for","paper_url":"https://arxiv.org/abs/2411.10564v2","paper_title":"Vision Eagle Attention: a new lens for advancing image classification","code":"https://github.com/MahmudulHasan11085/Vision-Eagle-Attention","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"StiDi-BP in R-CSNN","metrics":{"Accuracy":"92.8","Percentage error":"7.2"},"uses_additional_data":false,"paper_date":"2021-08-31","paper":"/paper/spike-time-displacement-based-error","paper_url":"https://arxiv.org/abs/2108.13621v1","paper_title":"Spike time displacement based error backpropagation in convolutional spiking neural networks","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"NeuPDE","metrics":{"Percentage error":"7.6"},"uses_additional_data":false,"paper_date":"2019-08-08","paper":"/paper/neupde-neural-network-based-ordinary-and","paper_url":"https://arxiv.org/abs/1908.03190v1","paper_title":"NeuPDE: Neural Network Based Ordinary and Partial Differential Equations for Modeling Time-Dependent Data","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"Star Algorithm on LeNet","metrics":{"Accuracy":"92.3","Percentage error":"7.7"},"uses_additional_data":false,"paper_date":"2022-06-01","paper":"/paper/star-algorithm-for-nn-ensembling","paper_url":"https://arxiv.org/abs/2206.00255v1","paper_title":"Star algorithm for NN ensembling","code":"https://github.com/mordiggian174/star-ensembling","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"ResNet-18","metrics":{"Accuracy":"92.28","Percentage error":"7.72"},"uses_additional_data":false,"paper_date":"2024-11-15","paper":"/paper/vision-eagle-attention-a-new-lens-for","paper_url":"https://arxiv.org/abs/2411.10564v2","paper_title":"Vision Eagle Attention: a new lens for advancing image classification","code":"https://github.com/MahmudulHasan11085/Vision-Eagle-Attention","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"SpeedyLiteVision Network","metrics":{"Accuracy":"91.60","Percentage error":"8.4","Trainable Parameters":"35978"},"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":16,"model":"CTM-8000 (Convolutional Tsetlin Machine)","metrics":{"Accuracy":"91.5","Percentage error":"8.5","Trainable Parameters":"527250"},"uses_additional_data":false,"paper_date":"2019-05-23","paper":"/paper/the-convolutional-tsetlin-machine","paper_url":"https://arxiv.org/abs/1905.09688v5","paper_title":"The Convolutional Tsetlin Machine","code":"https://github.com/cair/TsetlinMachine","n_code_links":9,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"Convolutional PMM (Parametric Matrix Model)","metrics":{"Accuracy":"90.89","Percentage error":"9.11","Trainable Parameters":"278280"},"uses_additional_data":false,"paper_date":"2024-01-22","paper":"/paper/parametric-matrix-models","paper_url":"https://arxiv.org/abs/2401.11694v6","paper_title":"Parametric Matrix Models","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":18,"model":"DENSER","metrics":{"Accuracy":"0.904","Percentage error":"9.6","Power consumption":"97.8"},"uses_additional_data":false,"paper_date":"2024-01-31","paper":"/paper/towards-physical-plausibility-in","paper_url":"https://arxiv.org/abs/2401.17733v1","paper_title":"Towards Physical Plausibility in Neuroevolution Systems","code":"https://github.com/rodriguesGabriel/energize","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"OTTT","metrics":{"Percentage error":"9.6"},"uses_additional_data":false,"paper_date":"2022-10-09","paper":"/paper/online-training-through-time-for-spiking","paper_url":"https://arxiv.org/abs/2210.04195v2","paper_title":"Online Training Through Time for Spiking Neural Networks","code":"https://github.com/pkuxmq/ottt-snn","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":20,"model":"ENERGIZE","metrics":{"Accuracy":"0.902","Percentage error":"9.8","Power consumption":"71.92"},"uses_additional_data":false,"paper_date":"2024-01-31","paper":"/paper/towards-physical-plausibility-in","paper_url":"https://arxiv.org/abs/2401.17733v1","paper_title":"Towards Physical Plausibility in Neuroevolution Systems","code":"https://github.com/rodriguesGabriel/energize","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"PMM (Parametric Matrix Model)","metrics":{"Accuracy":"88.58","Percentage error":"11.42","Trainable Parameters":"16744"},"uses_additional_data":false,"paper_date":"2024-01-22","paper":"/paper/parametric-matrix-models","paper_url":"https://arxiv.org/abs/2401.11694v6","paper_title":"Parametric Matrix Models","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":22,"model":"CTM-250 (Convolutional Tsetlin Machine)","metrics":{"Accuracy":"88.25","Percentage error":"11.75","Trainable Parameters":"31750"},"uses_additional_data":false,"paper_date":"2019-05-23","paper":"/paper/the-convolutional-tsetlin-machine","paper_url":"https://arxiv.org/abs/1905.09688v5","paper_title":"The Convolutional Tsetlin Machine","code":"https://github.com/cair/TsetlinMachine","n_code_links":9,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"GECCO","metrics":{"Accuracy":"88.09","Percentage error":"11.91"},"uses_additional_data":false,"paper_date":"2024-02-01","paper":"/paper/a-single-graph-convolution-is-all-you-need","paper_url":"https://arxiv.org/abs/2402.00564v6","paper_title":"A Single Graph Convolution Is All You Need: Efficient Grayscale Image Classification","code":"https://github.com/geccoproject/gecco","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"pFedBreD_ns_mg","metrics":{"Accuracy":"99.06"},"uses_additional_data":false,"paper_date":"2022-11-19","paper":"/paper/personalized-federated-learning-with-hidden","paper_url":"https://arxiv.org/abs/2211.10684v2","paper_title":"Personalized Federated Learning with Hidden Information on Personalized Prior","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":25,"model":"LR-Net","metrics":{"Accuracy":"95.03"},"uses_additional_data":false,"paper_date":"2022-07-19","paper":"/paper/a-block-based-convolutional-neural-network","paper_url":"https://arxiv.org/abs/2207.09531v5","paper_title":"LR-Net: A Block-based Convolutional Neural Network for Low-Resolution Image Classification","code":"https://github.com/AshkanGanj/Block-Based-ImageClassification-Architecture","n_code_links":1,"syntology":null},{"rank_in_archive_order":26,"model":"Tsetlin Machine Composites","metrics":{"Accuracy":"93.0"},"uses_additional_data":false,"paper_date":"2023-09-09","paper":"/paper/tmcomposites-plug-and-play-collaboration","paper_url":"https://arxiv.org/abs/2309.04801v2","paper_title":"TMComposites: Plug-and-Play Collaboration Between Specialized Tsetlin Machines","code":"https://github.com/cair/plug-and-play-collaboration-between-specialized-tsetlin-machines","n_code_links":1,"syntology":null},{"rank_in_archive_order":27,"model":"CNN+ Wilson-Cowan model RNN","metrics":{"Accuracy":"91.35"},"uses_additional_data":false,"paper_date":"2024-06-24","paper":"/paper/learning-in-wilson-cowan-model-for","paper_url":"https://arxiv.org/abs/2406.16453v2","paper_title":"Learning in Wilson-Cowan model for metapopulation","code":"https://github.com/raffaelemarino/learning_in_wilsoncowan","n_code_links":1,"syntology":null},{"rank_in_archive_order":28,"model":"FastSNN (CNN)","metrics":{"Accuracy":"90.57"},"uses_additional_data":false,"paper_date":"2022-05-30","paper":"/paper/accelerating-spiking-neural-network-training","paper_url":"https://arxiv.org/abs/2205.15286v2","paper_title":"Robust and accelerated single-spike spiking neural network training with applicability to challenging temporal tasks","code":"https://github.com/webstorms/fastsnn","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":29,"model":"FastSNN (MLP)","metrics":{"Accuracy":"89.05"},"uses_additional_data":false,"paper_date":"2022-05-30","paper":"/paper/accelerating-spiking-neural-network-training","paper_url":"https://arxiv.org/abs/2205.15286v2","paper_title":"Robust and accelerated single-spike spiking neural network training with applicability to challenging temporal tasks","code":"https://github.com/webstorms/fastsnn","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":30,"model":"Wilson-Cowan model RNN","metrics":{"Accuracy":"88.39"},"uses_additional_data":false,"paper_date":"2024-06-24","paper":"/paper/learning-in-wilson-cowan-model-for","paper_url":"https://arxiv.org/abs/2406.16453v2","paper_title":"Learning in Wilson-Cowan model for metapopulation","code":"https://github.com/raffaelemarino/learning_in_wilsoncowan","n_code_links":1,"syntology":null},{"rank_in_archive_order":31,"model":"Sparse Spiking Gradient Descent (CNN)","metrics":{"Accuracy":"86.7"},"uses_additional_data":false,"paper_date":"2021-05-18","paper":"/paper/sparse-spiking-gradient-descent","paper_url":"https://arxiv.org/abs/2105.08810v2","paper_title":"Sparse Spiking Gradient Descent","code":"https://github.com/npvoid/SparseSpikingBackprop","n_code_links":1,"syntology":null},{"rank_in_archive_order":32,"model":"SNNL-5","metrics":{"Accuracy":"84.4","NMI":"76.2"},"uses_additional_data":false,"paper_date":"2020-06-05","paper":"/paper/improving-k-means-clustering-performance-with","paper_url":"https://arxiv.org/abs/2006.04535v1","paper_title":"Improving k-Means Clustering Performance with Disentangled Internal Representations","code":"https://gitlab.com/afagarap/pt-snnl","n_code_links":1,"syntology":null},{"rank_in_archive_order":33,"model":"Continued fraction of straight lines","metrics":{"Accuracy":"84.12","NMI":"74.4","Trainable Parameters":"7870"},"uses_additional_data":false,"paper_date":"2024-12-16","paper":"/paper/real-valued-continued-fraction-of-straight","paper_url":"https://arxiv.org/abs/2412.16191v1","paper_title":"Real-valued continued fraction of straight lines","code":"https://github.com/grasshopper14/Continued-fraction-of-straight-lines/blob/main/continued_fraction_reg.py","n_code_links":1,"syntology":null},{"rank_in_archive_order":34,"model":"Sparse Spiking Gradient Descent (MLP)","metrics":{"Accuracy":"82.7"},"uses_additional_data":false,"paper_date":"2021-05-18","paper":"/paper/sparse-spiking-gradient-descent","paper_url":"https://arxiv.org/abs/2105.08810v2","paper_title":"Sparse Spiking Gradient Descent","code":"https://github.com/npvoid/SparseSpikingBackprop","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":6,"rows_with_any_sample_ran":2,"distinct_papers_with_graph_line":4,"distinct_papers_with_any_sample_ran":2,"samples_over_distinct_papers":{"n_ran":11,"n_unverified":3,"n_samples":14,"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":11,"n_unverified":6,"n_samples":17,"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"}}}