{"url":"/sota/sequential-image-classification-on-sequential-1","task":{"name":"Sequential Image Classification","url":"/task/sequential-image-classification","note":null},"dataset":{"name":"Sequential CIFAR-10","url":null},"category":"Computer Vision","categories":["Adversarial","Computer Vision"],"category_note":null,"description":"Sequential image classification is the task of classifying a sequence of images.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [TensorFlow-101](https://github.com/sjchoi86/tensorflow-101/blob/master/notebooks/rnn_mnist_simple.ipynb) )</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":["Unpermuted Accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Unpermuted Accuracy":"higher"}},"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":"MultiresConv","metrics":{"Unpermuted Accuracy":"93.15%"},"uses_additional_data":false,"paper_date":"2023-05-02","paper":"/paper/sequence-modeling-with-multiresolution","paper_url":"https://arxiv.org/abs/2305.01638v2","paper_title":"Sequence Modeling with Multiresolution Convolutional Memory","code":"https://github.com/thjashin/multires-conv","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":5,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"S4","metrics":{"Unpermuted Accuracy":"91.80%"},"uses_additional_data":false,"paper_date":"2021-10-31","paper":"/paper/efficiently-modeling-long-sequences-with-1","paper_url":"https://arxiv.org/abs/2111.00396v3","paper_title":"Efficiently Modeling Long Sequences with Structured State Spaces","code":"https://github.com/state-spaces/s4","n_code_links":8,"syntology":{"n_ran":28,"n_unverified":27,"n_samples":55,"n_pointer_only_licence":3}},{"rank_in_archive_order":3,"model":"LRU","metrics":{"Unpermuted Accuracy":"89.0"},"uses_additional_data":false,"paper_date":"2023-03-11","paper":"/paper/resurrecting-recurrent-neural-networks-for","paper_url":"https://arxiv.org/abs/2303.06349v1","paper_title":"Resurrecting Recurrent Neural Networks for Long Sequences","code":"https://github.com/Gothos/LRU-pytorch","n_code_links":11,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"SMPConv","metrics":{"Unpermuted Accuracy":"84.86%"},"uses_additional_data":false,"paper_date":"2023-04-05","paper":"/paper/smpconv-self-moving-point-representations-for","paper_url":"https://arxiv.org/abs/2304.02330v1","paper_title":"SMPConv: Self-moving Point Representations for Continuous Convolution","code":"https://github.com/sangnekim/smpconv","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":7,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"LSSL","metrics":{"Unpermuted Accuracy":"84.65%"},"uses_additional_data":false,"paper_date":"2021-10-26","paper":"/paper/combining-recurrent-convolutional-and","paper_url":"https://arxiv.org/abs/2110.13985v1","paper_title":"Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layers","code":"https://github.com/hazyresearch/state-spaces","n_code_links":2,"syntology":{"n_ran":19,"n_unverified":34,"n_samples":53,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"FlexTCN-6","metrics":{"Unpermuted Accuracy":"80.82%"},"uses_additional_data":false,"paper_date":"2021-10-15","paper":"/paper/flexconv-continuous-kernel-convolutions-with-1","paper_url":"https://arxiv.org/abs/2110.08059v3","paper_title":"FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizes","code":"https://github.com/rjbruin/flexconv","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"UR-GRU","metrics":{"Unpermuted Accuracy":"74.4%"},"uses_additional_data":false,"paper_date":"2019-10-22","paper":"/paper/improving-the-gating-mechanism-of-recurrent-1","paper_url":"https://arxiv.org/abs/1910.09890v2","paper_title":"Improving the Gating Mechanism of Recurrent Neural Networks","code":"https://github.com/aithlab/ImprovingGate","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"Trellis Network","metrics":{"Unpermuted Accuracy":"73.42%"},"uses_additional_data":false,"paper_date":"2018-10-15","paper":"/paper/trellis-networks-for-sequence-modeling","paper_url":"http://arxiv.org/abs/1810.06682v2","paper_title":"Trellis Networks for Sequence Modeling","code":"https://github.com/locuslab/trellisnet","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":6,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"Sparse Combo Net","metrics":{"Unpermuted Accuracy":"65.72"},"uses_additional_data":false,"paper_date":"2021-06-16","paper":"/paper/recursive-construction-of-stable-assemblies","paper_url":"https://arxiv.org/abs/2106.08928v6","paper_title":"RNNs of RNNs: Recursive Construction of Stable Assemblies of Recurrent Neural Networks","code":"https://github.com/ennisthemennis/sparse-combo-net","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"LipschitzRNN","metrics":{"Unpermuted Accuracy":"64.2"},"uses_additional_data":false,"paper_date":"2020-06-22","paper":"/paper/lipschitz-recurrent-neural-networks","paper_url":"https://arxiv.org/abs/2006.12070v3","paper_title":"Lipschitz Recurrent Neural Networks","code":"https://github.com/erichson/LipschitzRNN","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":11,"model":"CKCNN (1M)","metrics":{"Unpermuted Accuracy":"63.74%"},"uses_additional_data":false,"paper_date":"2021-02-04","paper":"/paper/ckconv-continuous-kernel-convolution-for","paper_url":"https://arxiv.org/abs/2102.02611v3","paper_title":"CKConv: Continuous Kernel Convolution For Sequential Data","code":"https://github.com/dwromero/ckconv","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"CKCNN (100k)","metrics":{"Unpermuted Accuracy":"62.25%"},"uses_additional_data":false,"paper_date":"2021-02-04","paper":"/paper/ckconv-continuous-kernel-convolution-for","paper_url":"https://arxiv.org/abs/2102.02611v3","paper_title":"CKConv: Continuous Kernel Convolution For Sequential Data","code":"https://github.com/dwromero/ckconv","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"Transformer (self-attention) (Trinh et al., 2018)","metrics":{"Unpermuted Accuracy":"62.2%"},"uses_additional_data":false,"paper_date":"2018-03-01","paper":"/paper/learning-longer-term-dependencies-in-rnns","paper_url":"http://arxiv.org/abs/1803.00144v3","paper_title":"Learning Longer-term Dependencies in RNNs with Auxiliary Losses","code":"https://github.com/younggyoseo/rnn-auxiliary-loss","n_code_links":1,"syntology":null}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"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":7,"rows_with_any_sample_ran":5,"distinct_papers_with_graph_line":7,"distinct_papers_with_any_sample_ran":5,"samples_over_distinct_papers":{"n_ran":54,"n_unverified":81,"n_samples":135,"n_pointer_only_licence":4,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":54,"n_unverified":81,"n_samples":135,"n_pointer_only_licence":4,"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"}}}