{"url":"/sota/long-tail-learning-on-cifar-100-lt-r-10","task":{"name":"Long-tail Learning","url":"/task/long-tail-learning","note":null},"dataset":{"name":"CIFAR-100-LT (ρ=10)","url":"/dataset/cifar-100"},"category":"Methodology","categories":["Methodology"],"category_note":null,"description":"Long-tailed learning, one of the most challenging problems in visual recognition, aims to train well-performing models from a large number of images that follow a long-tailed class distribution.","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":["Error Rate"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Error Rate":"lower"}},"counts":{"rows":31,"rows_with_code":25,"rows_with_paper_page":31,"rows_dated":31,"rows_using_additional_data":2},"rows":[{"rank_in_archive_order":1,"model":"LIFT (ViT-B/16, ImageNet-21K pre-training)","metrics":{"Error Rate":"8.7"},"uses_additional_data":false,"paper_date":"2023-09-18","paper":"/paper/parameter-efficient-long-tailed-recognition","paper_url":"https://arxiv.org/abs/2309.10019v3","paper_title":"Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts","code":"https://github.com/shijxcs/lift","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":3,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":2,"model":"LPT","metrics":{"Error Rate":"9"},"uses_additional_data":true,"paper_date":"2022-10-03","paper":"/paper/lpt-long-tailed-prompt-tuning-for-image","paper_url":"https://arxiv.org/abs/2210.01033v2","paper_title":"LPT: Long-tailed Prompt Tuning for Image Classification","code":"https://github.com/dongsky/lpt","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":3,"model":"VPT","metrics":{"Error Rate":"10.4"},"uses_additional_data":true,"paper_date":"2022-03-23","paper":"/paper/visual-prompt-tuning","paper_url":"https://arxiv.org/abs/2203.12119v2","paper_title":"Visual Prompt Tuning","code":"https://github.com/KMnP/vpt","n_code_links":6,"syntology":{"n_ran":17,"n_unverified":10,"n_samples":27,"n_pointer_only_licence":15}},{"rank_in_archive_order":4,"model":"LIFT (ViT-B/16, CLIP)","metrics":{"Error Rate":"15.1"},"uses_additional_data":false,"paper_date":"2023-09-18","paper":"/paper/parameter-efficient-long-tailed-recognition","paper_url":"https://arxiv.org/abs/2309.10019v3","paper_title":"Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts","code":"https://github.com/shijxcs/lift","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":3,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":5,"model":"GLMC+MaxNorm (ResNet-32, channel x4)","metrics":{"Error Rate":"25.72"},"uses_additional_data":false,"paper_date":"2023-05-15","paper":"/paper/global-and-local-mixture-consistency-1","paper_url":"https://arxiv.org/abs/2305.08661v1","paper_title":"Global and Local Mixture Consistency Cumulative Learning for Long-tailed Visual Recognitions","code":"https://github.com/ynu-yangpeng/GLMC","n_code_links":2,"syntology":{"n_ran":7,"n_unverified":3,"n_samples":10,"n_pointer_only_licence":5}},{"rank_in_archive_order":6,"model":"GLMC (ResNet-34, channel x4)","metrics":{"Error Rate":"26.53"},"uses_additional_data":false,"paper_date":"2023-05-15","paper":"/paper/global-and-local-mixture-consistency-1","paper_url":"https://arxiv.org/abs/2305.08661v1","paper_title":"Global and Local Mixture Consistency Cumulative Learning for Long-tailed Visual Recognitions","code":"https://github.com/ynu-yangpeng/GLMC","n_code_links":2,"syntology":{"n_ran":7,"n_unverified":3,"n_samples":10,"n_pointer_only_licence":5}},{"rank_in_archive_order":7,"model":"SURE(ResNet-32)","metrics":{"Error Rate":"26.76"},"uses_additional_data":false,"paper_date":"2024-03-01","paper":"/paper/sure-survey-recipes-for-building-reliable-and","paper_url":"https://arxiv.org/abs/2403.00543v1","paper_title":"SURE: SUrvey REcipes for building reliable and robust deep networks","code":"https://github.com/YutingLi0606/SURE","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":1,"n_samples":6,"n_pointer_only_licence":6}},{"rank_in_archive_order":8,"model":"PC","metrics":{"Error Rate":"30.88"},"uses_additional_data":false,"paper_date":"2023-02-01","paper":"/paper/learning-prototype-classifiers-for-long","paper_url":"https://arxiv.org/abs/2302.00491v3","paper_title":"Learning Prototype Classifiers for Long-Tailed Recognition","code":"https://github.com/saurabhsharma1993/prototype-classifier-ltr","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"LTR-weight-balancing","metrics":{"Error Rate":"31.33"},"uses_additional_data":false,"paper_date":"2022-03-27","paper":"/paper/long-tailed-recognition-via-weight-balancing","paper_url":"https://arxiv.org/abs/2203.14197v1","paper_title":"Long-Tailed Recognition via Weight Balancing","code":"https://github.com/shadealsha/ltr-weight-balancing","n_code_links":2,"syntology":{"n_ran":5,"n_unverified":9,"n_samples":14,"n_pointer_only_licence":13}},{"rank_in_archive_order":10,"model":"GML (ResNet-32)","metrics":{"Error Rate":"33.0"},"uses_additional_data":false,"paper_date":"2023-05-02","paper":"/paper/long-tailed-recognition-by-mutual-information","paper_url":"https://arxiv.org/abs/2305.01160v3","paper_title":"Long-Tailed Recognition by Mutual Information Maximization between Latent Features and Ground-Truth Labels","code":"https://github.com/bluecdm/Long-tailed-recognition","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"VS + ADRW + TLA","metrics":{"Error Rate":"34.41"},"uses_additional_data":false,"paper_date":"2023-10-07","paper":"/paper/a-unified-generalization-analysis-of-re","paper_url":"https://arxiv.org/abs/2310.04752","paper_title":"A Unified Generalization Analysis of Re-Weighting and Logit-Adjustment for Imbalanced Learning. paper with code","code":"https://github.com/wang22ti/DDC","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"Difficulty-Net","metrics":{"Error Rate":"34.78"},"uses_additional_data":false,"paper_date":"2022-09-07","paper":"/paper/difficulty-net-learning-to-predict-difficulty","paper_url":"https://arxiv.org/abs/2209.02960v1","paper_title":"Difficulty-Net: Learning to Predict Difficulty for Long-Tailed Recognition","code":"https://github.com/hitachi-rd-cv/Difficulty_Net","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"BCL+CUDA","metrics":{"Error Rate":"35.4"},"uses_additional_data":false,"paper_date":"2023-02-10","paper":"/paper/cuda-curriculum-of-data-augmentation-for-long","paper_url":"https://arxiv.org/abs/2302.05499v1","paper_title":"CUDA: Curriculum of Data Augmentation for Long-Tailed Recognition","code":"https://github.com/sumyeongahn/cuda_ltr","n_code_links":1,"syntology":{"n_ran":18,"n_unverified":2,"n_samples":20,"n_pointer_only_licence":20}},{"rank_in_archive_order":14,"model":"GLAG","metrics":{"Error Rate":"35.5"},"uses_additional_data":false,"paper_date":"2022-02-28","paper":"/paper/long-tailed-classification-with-gradual","paper_url":"https://arxiv.org/abs/2203.00452v1","paper_title":"Long-Tailed Classification with Gradual Balanced Loss and Adaptive Feature Generation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":15,"model":"LCReg","metrics":{"Error Rate":"35.8"},"uses_additional_data":false,"paper_date":"2022-06-02","paper":"/paper/long-tailed-recognition-by-learning-from-1","paper_url":"https://arxiv.org/abs/2206.01010v3","paper_title":"Long-tailed Recognition by Learning from Latent Categories","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":16,"model":"TADE","metrics":{"Error Rate":"36.4"},"uses_additional_data":false,"paper_date":"2021-07-20","paper":"/paper/test-agnostic-long-tailed-recognition-by-test","paper_url":"https://arxiv.org/abs/2107.09249v4","paper_title":"Self-Supervised Aggregation of Diverse Experts for Test-Agnostic Long-Tailed Recognition","code":"https://github.com/vanint/sade-agnosticlt","n_code_links":2,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"DRO-LT","metrics":{"Error Rate":"36.59"},"uses_additional_data":false,"paper_date":"2021-04-07","paper":"/paper/distributional-robustness-loss-for-long-tail","paper_url":"https://arxiv.org/abs/2104.03066v2","paper_title":"Distributional Robustness Loss for Long-tail Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":18,"model":"MiSLAS","metrics":{"Error Rate":"36.8"},"uses_additional_data":false,"paper_date":"2021-04-01","paper":"/paper/improving-calibration-for-long-tailed-1","paper_url":"https://arxiv.org/abs/2104.00466v1","paper_title":"Improving Calibration for Long-Tailed Recognition","code":"https://github.com/Jia-Research-Lab/MiSLAS","n_code_links":5,"syntology":{"n_ran":4,"n_unverified":5,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"SMC","metrics":{"Error Rate":"37.5"},"uses_additional_data":false,"paper_date":"2022-11-22","paper":"/paper/supervised-contrastive-learning-on-blended","paper_url":"https://arxiv.org/abs/2211.11938v1","paper_title":"Supervised Contrastive Learning on Blended Images for Long-tailed Recognition","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":20,"model":"Hybrid-PSC","metrics":{"Error Rate":"37.63"},"uses_additional_data":false,"paper_date":"2021-03-26","paper":"/paper/contrastive-learning-based-hybrid-networks","paper_url":"https://arxiv.org/abs/2103.14267v1","paper_title":"Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":21,"model":"LADE","metrics":{"Error Rate":"38.3"},"uses_additional_data":false,"paper_date":"2020-12-01","paper":"/paper/disentangling-label-distribution-for-long","paper_url":"https://arxiv.org/abs/2012.00321v2","paper_title":"Disentangling Label Distribution for Long-tailed Visual Recognition","code":"https://github.com/hyperconnect/LADE","n_code_links":2,"syntology":null},{"rank_in_archive_order":22,"model":"RIDE + CMO + Curvature Regularization","metrics":{"Error Rate":"38.60"},"uses_additional_data":false,"paper_date":"2023-03-22","paper":"/paper/curvature-balanced-feature-manifold-learning","paper_url":"https://arxiv.org/abs/2303.12307v7","paper_title":"Predicting and Enhancing the Fairness of DNNs with the Curvature of Perceptual Manifolds","code":"https://github.com/mayanbiao1234/geometric-metrics-for-perceptual-manifolds","n_code_links":2,"syntology":null},{"rank_in_archive_order":23,"model":"MetaSAug-LDAM","metrics":{"Error Rate":"38.72"},"uses_additional_data":false,"paper_date":"2021-03-23","paper":"/paper/metasaug-meta-semantic-augmentation-for-long","paper_url":"https://arxiv.org/abs/2103.12579v3","paper_title":"MetaSAug: Meta Semantic Augmentation for Long-Tailed Visual Recognition","code":"https://github.com/BIT-DA/MetaSAug","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":24,"model":"UniMix+Bayias (ResNet-32)","metrics":{"Error Rate":"38.75"},"uses_additional_data":false,"paper_date":"2021-11-06","paper":"/paper/towards-calibrated-model-for-long-tailed","paper_url":"https://arxiv.org/abs/2111.03874v1","paper_title":"Towards Calibrated Model for Long-Tailed Visual Recognition from Prior Perspective","code":"https://github.com/XuZhengzhuo/Prior-LT","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":7,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":25,"model":"CBD+TailCalibX","metrics":{"Error Rate":"38.87"},"uses_additional_data":false,"paper_date":"2021-11-10","paper":"/paper/feature-generation-for-long-tail","paper_url":"https://arxiv.org/abs/2111.05956v1","paper_title":"Feature Generation for Long-tail Classification","code":"https://github.com/rahulvigneswaran/tailcalibx","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":5,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"ELP","metrics":{"Error Rate":"40.9"},"uses_additional_data":false,"paper_date":"2022-01-01","paper":"/paper/a-simple-episodic-linear-probe-improves","paper_url":"http://openaccess.thecvf.com//content/CVPR2022/html/Liang_A_Simple_Episodic_Linear_Probe_Improves_Visual_Recognition_in_the_CVPR_2022_paper.html","paper_title":"A Simple Episodic Linear Probe Improves Visual Recognition in the Wild","code":"https://github.com/JDAI-CV/DCL","n_code_links":2,"syntology":null},{"rank_in_archive_order":27,"model":"LDAM-DRW + SSP","metrics":{"Error Rate":"41.09"},"uses_additional_data":false,"paper_date":"2020-06-13","paper":"/paper/rethinking-the-value-of-labels-for-improving","paper_url":"https://arxiv.org/abs/2006.07529v2","paper_title":"Rethinking the Value of Labels for Improving Class-Imbalanced Learning","code":"https://github.com/YyzHarry/imbalanced-semi-self","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":28,"model":"smDRAGON","metrics":{"Error Rate":"41.23"},"uses_additional_data":false,"paper_date":"2020-04-05","paper":"/paper/long-tail-learning-with-attributes","paper_url":"https://arxiv.org/abs/2004.02235v4","paper_title":"From Generalized zero-shot learning to long-tail with class descriptors","code":"https://github.com/dvirsamuel/DRAGON","n_code_links":1,"syntology":null},{"rank_in_archive_order":29,"model":"CDB-loss","metrics":{"Error Rate":"41.26"},"uses_additional_data":false,"paper_date":"2020-10-05","paper":"/paper/class-wise-difficulty-balanced-loss-for","paper_url":"https://arxiv.org/abs/2010.01824v1","paper_title":"Class-Wise Difficulty-Balanced Loss for Solving Class-Imbalance","code":"https://github.com/hitachi-rd-cv/CDB-loss","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":30,"model":"LDAM-DRW","metrics":{"Error Rate":"41.29"},"uses_additional_data":false,"paper_date":"2019-06-18","paper":"/paper/learning-imbalanced-datasets-with-label","paper_url":"https://arxiv.org/abs/1906.07413v2","paper_title":"Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss","code":"https://github.com/kaidic/LDAM-DRW","n_code_links":7,"syntology":{"n_ran":2,"n_unverified":9,"n_samples":11,"n_pointer_only_licence":3}},{"rank_in_archive_order":31,"model":"CE-DRW-IC","metrics":{"Error Rate":"41.4"},"uses_additional_data":false,"paper_date":"2020-10-22","paper":"/paper/posterior-re-calibration-for-imbalanced","paper_url":"https://arxiv.org/abs/2010.11820v1","paper_title":"Posterior Re-calibration for Imbalanced Datasets","code":null,"n_code_links":0,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}}],"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":19,"rows_with_any_sample_ran":16,"distinct_papers_with_graph_line":17,"distinct_papers_with_any_sample_ran":14,"samples_over_distinct_papers":{"n_ran":74,"n_unverified":60,"n_samples":134,"n_pointer_only_licence":74,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":86,"n_unverified":66,"n_samples":152,"n_pointer_only_licence":87,"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"}}}