{"url":"/sota/semi-supervised-image-classification-on-1","task":{"name":"Semi-Supervised Image Classification","url":"/task/semi-supervised-image-classification","note":null},"dataset":{"name":"ImageNet - 1% labeled data","url":"/dataset/imagenet"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"Semi-supervised image classification leverages unlabelled data as well as labelled data to increase classification performance.\r\n\r\nYou may want to read some blog posts to get an overview before reading the papers and checking the leaderboards:\r\n\r\n- [An overview of proxy-label approaches for semi-supervised learning](https://ruder.io/semi-supervised/) - Sebastian Ruder\r\n- [Semi-Supervised Learning in Computer Vision](https://amitness.com/2020/07/semi-supervised-learning/) - Amit Chaudhary\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Self-Supervised Semi-Supervised Learning](https://arxiv.org/pdf/1905.03670v2.pdf) )</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":["Top 1 Accuracy","Top 5 Accuracy","Number of params"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Top 1 Accuracy":"higher","Top 5 Accuracy":"higher","Number of params":"lower"}},"counts":{"rows":65,"rows_with_code":59,"rows_with_paper_page":64,"rows_dated":64,"rows_using_additional_data":1},"rows":[{"rank_in_archive_order":1,"model":"DHO (ViT-Large)","metrics":{"Top 1 Accuracy":"84.6%"},"uses_additional_data":false,"paper_date":"2025-05-12","paper":"/paper/simple-semi-supervised-knowledge-distillation","paper_url":"https://arxiv.org/abs/2505.07675v1","paper_title":"Simple Semi-supervised Knowledge Distillation from Vision-Language Models via $\\mathbf{\\texttt{D}}$ual-$\\mathbf{\\texttt{H}}$ead $\\mathbf{\\texttt{O}}$ptimization","code":"https://github.com/erjui/DHO","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"REACT (ViT-Large)","metrics":{"Top 1 Accuracy":"81.6%"},"uses_additional_data":true,"paper_date":"2023-01-17","paper":"/paper/learning-customized-visual-models-with","paper_url":"https://arxiv.org/abs/2301.07094v1","paper_title":"Learning Customized Visual Models with Retrieval-Augmented Knowledge","code":"https://github.com/microsoft/react","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":11,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"DHO (ViT-Base)","metrics":{"Top 1 Accuracy":"81.6%"},"uses_additional_data":false,"paper_date":"2025-05-12","paper":"/paper/simple-semi-supervised-knowledge-distillation","paper_url":"https://arxiv.org/abs/2505.07675v1","paper_title":"Simple Semi-supervised Knowledge Distillation from Vision-Language Models via $\\mathbf{\\texttt{D}}$ual-$\\mathbf{\\texttt{H}}$ead $\\mathbf{\\texttt{O}}$ptimization","code":"https://github.com/erjui/DHO","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"Meta Co-Training","metrics":{"Top 1 Accuracy":"80.7%"},"uses_additional_data":false,"paper_date":"2023-11-29","paper":"/paper/meta-co-training-two-views-are-better-than","paper_url":"https://arxiv.org/abs/2311.18083v4","paper_title":"Meta Co-Training: Two Views are Better than One","code":"https://github.com/jayrothenberger/meta-co-training","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":5,"model":"Semi-SST (ViT-Huge)","metrics":{"Top 1 Accuracy":"80.7%"},"uses_additional_data":false,"paper_date":"2025-05-31","paper":"/paper/sst-self-training-with-self-adaptive-1","paper_url":"https://arxiv.org/abs/2506.00467v1","paper_title":"SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"Super-SST (ViT-Huge)","metrics":{"Top 1 Accuracy":"80.3%"},"uses_additional_data":false,"paper_date":"2025-05-31","paper":"/paper/sst-self-training-with-self-adaptive-1","paper_url":"https://arxiv.org/abs/2506.00467v1","paper_title":"SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"Semi-ViT (ViT-Huge)","metrics":{"Top 1 Accuracy":"80%","Top 5 Accuracy":"93.1"},"uses_additional_data":false,"paper_date":"2022-08-11","paper":"/paper/semi-supervised-vision-transformers-at-scale","paper_url":"https://arxiv.org/abs/2208.05688v1","paper_title":"Semi-supervised Vision Transformers at Scale","code":"https://github.com/amazon-science/semi-vit","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"Semi-ViT (ViT-Large)","metrics":{"Top 1 Accuracy":"77.3%"},"uses_additional_data":false,"paper_date":"2022-08-11","paper":"/paper/semi-supervised-vision-transformers-at-scale","paper_url":"https://arxiv.org/abs/2208.05688v1","paper_title":"Semi-supervised Vision Transformers at Scale","code":"https://github.com/amazon-science/semi-vit","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"Super-SST (ViT-Small distilled)","metrics":{"Top 1 Accuracy":"76.9%"},"uses_additional_data":false,"paper_date":"2025-05-31","paper":"/paper/sst-self-training-with-self-adaptive-1","paper_url":"https://arxiv.org/abs/2506.00467v1","paper_title":"SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":10,"model":"SimCLRv2 self-distilled (ResNet-152 x3, SK)","metrics":{"Top 1 Accuracy":"76.6%","Top 5 Accuracy":"93.4%"},"uses_additional_data":false,"paper_date":"2020-06-17","paper":"/paper/big-self-supervised-models-are-strong-semi","paper_url":"https://arxiv.org/abs/2006.10029v2","paper_title":"Big Self-Supervised Models are Strong Semi-Supervised Learners","code":"https://github.com/google-research/simclr","n_code_links":9,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"SimCLRv2 distilled (ResNet-50 x2, SK)","metrics":{"Top 1 Accuracy":"75.9%","Top 5 Accuracy":"93.0%"},"uses_additional_data":false,"paper_date":"2020-06-17","paper":"/paper/big-self-supervised-models-are-strong-semi","paper_url":"https://arxiv.org/abs/2006.10029v2","paper_title":"Big Self-Supervised Models are Strong Semi-Supervised Learners","code":"https://github.com/google-research/simclr","n_code_links":9,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"MSN (ViT-B/4)","metrics":{"Top 1 Accuracy":"75.7%"},"uses_additional_data":false,"paper_date":"2022-04-14","paper":"/paper/masked-siamese-networks-for-label-efficient","paper_url":"https://arxiv.org/abs/2204.07141v1","paper_title":"Masked Siamese Networks for Label-Efficient Learning","code":"https://github.com/lightly-ai/lightly","n_code_links":2,"syntology":{"n_ran":4,"n_unverified":0,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"SimCLRv2 (ResNet-152 x3, SK)","metrics":{"Top 1 Accuracy":"74.9%","Top 5 Accuracy":"92.3%"},"uses_additional_data":false,"paper_date":"2020-06-17","paper":"/paper/big-self-supervised-models-are-strong-semi","paper_url":"https://arxiv.org/abs/2006.10029v2","paper_title":"Big Self-Supervised Models are Strong Semi-Supervised Learners","code":"https://github.com/google-research/simclr","n_code_links":9,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":14,"model":"SimCLRv2 distilled (ResNet-50)","metrics":{"Top 1 Accuracy":"73.9%","Top 5 Accuracy":"91.5%"},"uses_additional_data":false,"paper_date":"2020-06-17","paper":"/paper/big-self-supervised-models-are-strong-semi","paper_url":"https://arxiv.org/abs/2006.10029v2","paper_title":"Big Self-Supervised Models are Strong Semi-Supervised Learners","code":"https://github.com/google-research/simclr","n_code_links":9,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"SimMatchV2 (ResNet-50)","metrics":{"Top 1 Accuracy":"71.9%"},"uses_additional_data":false,"paper_date":"2023-08-13","paper":"/paper/simmatchv2-semi-supervised-learning-with","paper_url":"https://arxiv.org/abs/2308.06692v1","paper_title":"SimMatchV2: Semi-Supervised Learning with Graph Consistency","code":"https://github.com/kylezheng1997/simmatch","n_code_links":2,"syntology":null},{"rank_in_archive_order":16,"model":"Semi-SST (ViT-Small)","metrics":{"Top 1 Accuracy":"71.4%"},"uses_additional_data":false,"paper_date":"2025-05-31","paper":"/paper/sst-self-training-with-self-adaptive-1","paper_url":"https://arxiv.org/abs/2506.00467v1","paper_title":"SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":17,"model":"DebiasPL (ResNet-50)","metrics":{"Top 1 Accuracy":"71.3%"},"uses_additional_data":false,"paper_date":"2022-01-05","paper":"/paper/debiased-learning-from-naturally-imbalanced","paper_url":"https://arxiv.org/abs/2201.01490v2","paper_title":"Debiased Learning from Naturally Imbalanced Pseudo-Labels","code":"https://github.com/frank-xwang/debiased-pseudo-labeling","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":18,"model":"BYOL (ResNet-200 x2)","metrics":{"Top 1 Accuracy":"71.2%","Top 5 Accuracy":"89.5%"},"uses_additional_data":false,"paper_date":"2020-06-13","paper":"/paper/bootstrap-your-own-latent-a-new-approach-to","paper_url":"https://arxiv.org/abs/2006.07733v3","paper_title":"Bootstrap your own latent: A new approach to self-supervised Learning","code":"https://github.com/deepmind/deepmind-research/tree/master/byol","n_code_links":31,"syntology":{"n_ran":62,"n_unverified":17,"n_samples":79,"n_pointer_only_licence":46}},{"rank_in_archive_order":19,"model":"Semi-ViT (ViT-Base)","metrics":{"Top 1 Accuracy":"71%"},"uses_additional_data":false,"paper_date":"2022-08-11","paper":"/paper/semi-supervised-vision-transformers-at-scale","paper_url":"https://arxiv.org/abs/2208.05688v1","paper_title":"Semi-supervised Vision Transformers at Scale","code":"https://github.com/amazon-science/semi-vit","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"Super-SST (ViT-Small)","metrics":{"Top 1 Accuracy":"70.4%"},"uses_additional_data":false,"paper_date":"2025-05-31","paper":"/paper/sst-self-training-with-self-adaptive-1","paper_url":"https://arxiv.org/abs/2506.00467v1","paper_title":"SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":21,"model":"PAWS (ResNet-50 4x)","metrics":{"Top 1 Accuracy":"69.9%"},"uses_additional_data":false,"paper_date":"2021-04-28","paper":"/paper/semi-supervised-learning-of-visual-features","paper_url":"https://arxiv.org/abs/2104.13963v3","paper_title":"Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples","code":"https://github.com/facebookresearch/suncet","n_code_links":4,"syntology":{"n_ran":7,"n_unverified":14,"n_samples":21,"n_pointer_only_licence":3}},{"rank_in_archive_order":22,"model":"PAWS (ResNet-50 2x)","metrics":{"Top 1 Accuracy":"69.6%"},"uses_additional_data":false,"paper_date":"2021-04-28","paper":"/paper/semi-supervised-learning-of-visual-features","paper_url":"https://arxiv.org/abs/2104.13963v3","paper_title":"Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples","code":"https://github.com/facebookresearch/suncet","n_code_links":4,"syntology":{"n_ran":7,"n_unverified":14,"n_samples":21,"n_pointer_only_licence":3}},{"rank_in_archive_order":23,"model":"BYOL (ResNet-50 x4)","metrics":{"Top 1 Accuracy":"69.1%","Top 5 Accuracy":"87.9%"},"uses_additional_data":false,"paper_date":"2020-06-13","paper":"/paper/bootstrap-your-own-latent-a-new-approach-to","paper_url":"https://arxiv.org/abs/2006.07733v3","paper_title":"Bootstrap your own latent: A new approach to self-supervised Learning","code":"https://github.com/deepmind/deepmind-research/tree/master/byol","n_code_links":31,"syntology":{"n_ran":62,"n_unverified":17,"n_samples":79,"n_pointer_only_licence":46}},{"rank_in_archive_order":24,"model":"SimMatch + EPASS (ResNet-50)","metrics":{"Top 1 Accuracy":"68.6%","Top 5 Accuracy":"87.6"},"uses_additional_data":false,"paper_date":"2023-10-24","paper":"/paper/debiasing-calibrating-and-improving-semi","paper_url":"https://arxiv.org/abs/2310.15764v1","paper_title":"Debiasing, calibrating, and improving Semi-supervised Learning performance via simple Ensemble Projector","code":"https://github.com/beandkay/epass","n_code_links":1,"syntology":null},{"rank_in_archive_order":25,"model":"CoMatch + EPASS (ResNet-50)","metrics":{"Top 1 Accuracy":"67.4%","Top 5 Accuracy":"87.3"},"uses_additional_data":false,"paper_date":"2023-10-24","paper":"/paper/debiasing-calibrating-and-improving-semi","paper_url":"https://arxiv.org/abs/2310.15764v1","paper_title":"Debiasing, calibrating, and improving Semi-supervised Learning performance via simple Ensemble Projector","code":"https://github.com/beandkay/epass","n_code_links":1,"syntology":null},{"rank_in_archive_order":26,"model":"TWIST (ResNet-50 x2)","metrics":{"Top 1 Accuracy":"67.2%","Top 5 Accuracy":"88.2%"},"uses_additional_data":false,"paper_date":"2021-10-14","paper":"/paper/self-supervised-learning-by-estimating-twin-1","paper_url":"https://arxiv.org/abs/2110.07402v4","paper_title":"Self-Supervised Learning by Estimating Twin Class Distributions","code":"https://github.com/bytedance/TWIST","n_code_links":2,"syntology":{"n_ran":5,"n_unverified":10,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"SimMatch (ResNet-50)","metrics":{"Top 1 Accuracy":"67.2%"},"uses_additional_data":false,"paper_date":"2022-03-14","paper":"/paper/simmatch-semi-supervised-learning-with","paper_url":"https://arxiv.org/abs/2203.06915v2","paper_title":"SimMatch: Semi-supervised Learning with Similarity Matching","code":"https://github.com/kylezheng1997/simmatch","n_code_links":1,"syntology":null},{"rank_in_archive_order":28,"model":"CoMatch (w. MoCo v2)","metrics":{"Top 1 Accuracy":"67.1%","Top 5 Accuracy":"87.1%"},"uses_additional_data":false,"paper_date":"2020-11-23","paper":"/paper/comatch-semi-supervised-learning-with","paper_url":"https://arxiv.org/abs/2011.11183v2","paper_title":"CoMatch: Semi-supervised Learning with Contrastive Graph Regularization","code":"https://github.com/salesforce/CoMatch","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":29,"model":"PAWS (ResNet-50)","metrics":{"Top 1 Accuracy":"66.5%"},"uses_additional_data":false,"paper_date":"2021-04-28","paper":"/paper/semi-supervised-learning-of-visual-features","paper_url":"https://arxiv.org/abs/2104.13963v3","paper_title":"Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples","code":"https://github.com/facebookresearch/suncet","n_code_links":4,"syntology":{"n_ran":7,"n_unverified":14,"n_samples":21,"n_pointer_only_licence":3}},{"rank_in_archive_order":30,"model":"SimCLRv2 (ResNet-50 ×2)","metrics":{"Top 1 Accuracy":"66.3%","Top 5 Accuracy":"87.4%"},"uses_additional_data":false,"paper_date":"2020-06-17","paper":"/paper/big-self-supervised-models-are-strong-semi","paper_url":"https://arxiv.org/abs/2006.10029v2","paper_title":"Big Self-Supervised Models are Strong Semi-Supervised Learners","code":"https://github.com/google-research/simclr","n_code_links":9,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":31,"model":"WCL (ResNet-50)","metrics":{"Top 1 Accuracy":"65.0%","Top 5 Accuracy":"86.3%"},"uses_additional_data":false,"paper_date":"2021-10-10","paper":"/paper/weakly-supervised-contrastive-learning-1","paper_url":"https://arxiv.org/abs/2110.04770v1","paper_title":"Weakly Supervised Contrastive Learning","code":"https://github.com/KyleZheng1997/WCL","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":32,"model":"FNC (ResNet-50)","metrics":{"Top 1 Accuracy":"63.7%","Top 5 Accuracy":"85.3%"},"uses_additional_data":false,"paper_date":"2020-11-23","paper":"/paper/boosting-contrastive-self-supervised-learning","paper_url":"https://arxiv.org/abs/2011.11765v2","paper_title":"Boosting Contrastive Self-Supervised Learning with False Negative Cancellation","code":"https://github.com/google-research/fnc","n_code_links":1,"syntology":null},{"rank_in_archive_order":33,"model":"SimCLR (ResNet-50 4×)","metrics":{"Top 1 Accuracy":"63.0%","Top 5 Accuracy":"85.8%"},"uses_additional_data":false,"paper_date":"2020-02-13","paper":"/paper/a-simple-framework-for-contrastive-learning","paper_url":"https://arxiv.org/abs/2002.05709v3","paper_title":"A Simple Framework for Contrastive Learning of Visual Representations","code":"https://github.com/tensorflow/models/tree/master/official/vision/beta/projects/simclr","n_code_links":96,"syntology":{"n_ran":79,"n_unverified":58,"n_samples":137,"n_pointer_only_licence":52}},{"rank_in_archive_order":34,"model":"FixMatch-EMAN","metrics":{"Top 1 Accuracy":"63%"},"uses_additional_data":false,"paper_date":"2021-01-21","paper":"/paper/exponential-moving-average-normalization-for","paper_url":"https://arxiv.org/abs/2101.08482v2","paper_title":"Exponential Moving Average Normalization for Self-supervised and Semi-supervised Learning","code":"https://github.com/amazon-research/exponential-moving-average-normalization","n_code_links":1,"syntology":null},{"rank_in_archive_order":35,"model":"SEER (RegNet10B)","metrics":{"Top 1 Accuracy":"62.4%"},"uses_additional_data":false,"paper_date":"2022-02-16","paper":"/paper/vision-models-are-more-robust-and-fair-when","paper_url":"https://arxiv.org/abs/2202.08360v2","paper_title":"Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision","code":"https://github.com/facebookresearch/vissl","n_code_links":1,"syntology":null},{"rank_in_archive_order":36,"model":"BYOL (ResNet-50 x2)","metrics":{"Top 1 Accuracy":"62.2%","Top 5 Accuracy":"84.1%"},"uses_additional_data":false,"paper_date":"2020-06-13","paper":"/paper/bootstrap-your-own-latent-a-new-approach-to","paper_url":"https://arxiv.org/abs/2006.07733v3","paper_title":"Bootstrap your own latent: A new approach to self-supervised Learning","code":"https://github.com/deepmind/deepmind-research/tree/master/byol","n_code_links":31,"syntology":{"n_ran":62,"n_unverified":17,"n_samples":79,"n_pointer_only_licence":46}},{"rank_in_archive_order":37,"model":"iBOT (ViT-S/16)","metrics":{"Top 1 Accuracy":"61.9%"},"uses_additional_data":false,"paper_date":"2021-11-15","paper":"/paper/ibot-image-bert-pre-training-with-online","paper_url":"https://arxiv.org/abs/2111.07832v3","paper_title":"iBOT: Image BERT Pre-Training with Online Tokenizer","code":"https://github.com/bytedance/ibot","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":38,"model":"SEER Large (RegNetY-256GF)","metrics":{"Top 1 Accuracy":"60.5%"},"uses_additional_data":false,"paper_date":"2021-03-02","paper":"/paper/self-supervised-pretraining-of-visual","paper_url":"https://arxiv.org/abs/2103.01988v2","paper_title":"Self-supervised Pretraining of Visual Features in the Wild","code":"https://github.com/facebookresearch/vissl","n_code_links":1,"syntology":null},{"rank_in_archive_order":39,"model":"SemiReward","metrics":{"Top 1 Accuracy":"59.64%"},"uses_additional_data":false,"paper_date":"2023-10-04","paper":"/paper/semireward-a-general-reward-model-for-semi","paper_url":"https://arxiv.org/abs/2310.03013v2","paper_title":"SemiReward: A General Reward Model for Semi-supervised Learning","code":"https://github.com/Westlake-AI/SemiReward","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":1,"n_samples":5,"n_pointer_only_licence":1}},{"rank_in_archive_order":40,"model":"SimCLR (ResNet-50 2×)","metrics":{"Top 1 Accuracy":"58.5%","Top 5 Accuracy":"83.0%"},"uses_additional_data":false,"paper_date":"2020-02-13","paper":"/paper/a-simple-framework-for-contrastive-learning","paper_url":"https://arxiv.org/abs/2002.05709v3","paper_title":"A Simple Framework for Contrastive Learning of Visual Representations","code":"https://github.com/tensorflow/models/tree/master/official/vision/beta/projects/simclr","n_code_links":96,"syntology":{"n_ran":79,"n_unverified":58,"n_samples":137,"n_pointer_only_licence":52}},{"rank_in_archive_order":41,"model":"RELICv2","metrics":{"Top 1 Accuracy":"58.1%","Top 5 Accuracy":"81.3"},"uses_additional_data":false,"paper_date":"2022-01-13","paper":"/paper/pushing-the-limits-of-self-supervised-resnets","paper_url":"https://arxiv.org/abs/2201.05119v2","paper_title":"Pushing the limits of self-supervised ResNets: Can we outperform supervised learning without labels on ImageNet?","code":"https://github.com/google-deepmind/relicv2","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":14,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":42,"model":"SimCLRv2 (ResNet-50)","metrics":{"Top 1 Accuracy":"57.9%","Top 5 Accuracy":"82.5%"},"uses_additional_data":false,"paper_date":"2020-06-17","paper":"/paper/big-self-supervised-models-are-strong-semi","paper_url":"https://arxiv.org/abs/2006.10029v2","paper_title":"Big Self-Supervised Models are Strong Semi-Supervised Learners","code":"https://github.com/google-research/simclr","n_code_links":9,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":43,"model":"SEER Small (RegNetY-128GF)","metrics":{"Top 1 Accuracy":"57.5%"},"uses_additional_data":false,"paper_date":"2021-03-02","paper":"/paper/self-supervised-pretraining-of-visual","paper_url":"https://arxiv.org/abs/2103.01988v2","paper_title":"Self-supervised Pretraining of Visual Features in the Wild","code":"https://github.com/facebookresearch/vissl","n_code_links":1,"syntology":null},{"rank_in_archive_order":44,"model":"NNCLR (ResNet-50)","metrics":{"Top 1 Accuracy":"56.4%","Top 5 Accuracy":"80.7"},"uses_additional_data":false,"paper_date":"2021-04-29","paper":"/paper/with-a-little-help-from-my-friends-nearest","paper_url":"https://arxiv.org/abs/2104.14548v2","paper_title":"With a Little Help from My Friends: Nearest-Neighbor Contrastive Learning of Visual Representations","code":"https://github.com/lightly-ai/lightly","n_code_links":4,"syntology":{"n_ran":4,"n_unverified":1,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":45,"model":"I-VNE+ (ResNet-50)","metrics":{"Top 1 Accuracy":"55.8","Top 5 Accuracy":"81.0"},"uses_additional_data":false,"paper_date":"2023-04-04","paper":"/paper/vne-an-effective-method-for-improving-deep","paper_url":"https://arxiv.org/abs/2304.01434v1","paper_title":"VNE: An Effective Method for Improving Deep Representation by Manipulating Eigenvalue Distribution","code":"https://github.com/jaeill/CVPR23-VNE","n_code_links":1,"syntology":null},{"rank_in_archive_order":46,"model":"Barlow Twins (ResNet-50)","metrics":{"Top 1 Accuracy":"55%","Top 5 Accuracy":"79.2"},"uses_additional_data":false,"paper_date":"2021-03-04","paper":"/paper/barlow-twins-self-supervised-learning-via","paper_url":"https://arxiv.org/abs/2103.03230v3","paper_title":"Barlow Twins: Self-Supervised Learning via Redundancy Reduction","code":"https://github.com/lightly-ai/lightly","n_code_links":24,"syntology":{"n_ran":21,"n_unverified":5,"n_samples":26,"n_pointer_only_licence":10}},{"rank_in_archive_order":47,"model":"VICREG (Resnet-50)","metrics":{"Top 1 Accuracy":"54.8%","Top 5 Accuracy":"79.4%"},"uses_additional_data":false,"paper_date":"2021-05-11","paper":"/paper/vicreg-variance-invariance-covariance","paper_url":"https://arxiv.org/abs/2105.04906v3","paper_title":"VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning","code":"https://github.com/lightly-ai/lightly","n_code_links":6,"syntology":{"n_ran":11,"n_unverified":6,"n_samples":17,"n_pointer_only_licence":1}},{"rank_in_archive_order":48,"model":"SwAV (ResNet-50)","metrics":{"Top 1 Accuracy":"53.9%","Top 5 Accuracy":"78.5"},"uses_additional_data":false,"paper_date":"2020-06-17","paper":"/paper/unsupervised-learning-of-visual-features-by","paper_url":"https://arxiv.org/abs/2006.09882v5","paper_title":"Unsupervised Learning of Visual Features by Contrasting Cluster Assignments","code":"https://github.com/open-mmlab/mmdetection","n_code_links":18,"syntology":{"n_ran":13,"n_unverified":4,"n_samples":17,"n_pointer_only_licence":6}},{"rank_in_archive_order":49,"model":"BYOL (ResNet-50)","metrics":{"Top 1 Accuracy":"53.2%","Top 5 Accuracy":"78.4%"},"uses_additional_data":false,"paper_date":"2020-06-13","paper":"/paper/bootstrap-your-own-latent-a-new-approach-to","paper_url":"https://arxiv.org/abs/2006.07733v3","paper_title":"Bootstrap your own latent: A new approach to self-supervised Learning","code":"https://github.com/deepmind/deepmind-research/tree/master/byol","n_code_links":31,"syntology":{"n_ran":62,"n_unverified":17,"n_samples":79,"n_pointer_only_licence":46}},{"rank_in_archive_order":50,"model":"CPC v2 (ResNet-161)","metrics":{"Top 1 Accuracy":"52.7%","Top 5 Accuracy":"77.9%"},"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":51,"model":"SynCo (ResNet-50) 800ep","metrics":{"Number of params":"24M","Top 1 Accuracy":"50.8%","Top 5 Accuracy":"77.5%"},"uses_additional_data":false,"paper_date":"2024-10-03","paper":"/paper/synco-synthetic-hard-negatives-in-contrastive","paper_url":"https://arxiv.org/abs/2410.02401v5","paper_title":"SynCo: Synthetic Hard Negatives in Contrastive Learning for Better Unsupervised Visual Representations","code":"https://github.com/giakoumoglou/synco","n_code_links":1,"syntology":null},{"rank_in_archive_order":52,"model":"SimCLR (ResNet-50)","metrics":{"Top 1 Accuracy":"48.3%","Top 5 Accuracy":"75.5%"},"uses_additional_data":false,"paper_date":"2020-02-13","paper":"/paper/a-simple-framework-for-contrastive-learning","paper_url":"https://arxiv.org/abs/2002.05709v3","paper_title":"A Simple Framework for Contrastive Learning of Visual Representations","code":"https://github.com/tensorflow/models/tree/master/official/vision/beta/projects/simclr","n_code_links":96,"syntology":{"n_ran":79,"n_unverified":58,"n_samples":137,"n_pointer_only_licence":52}},{"rank_in_archive_order":53,"model":"SCAN (ResNet-50|Unsupervised)","metrics":{"Top 1 Accuracy":"39.90%","Top 5 Accuracy":"60.0%"},"uses_additional_data":false,"paper_date":"2020-05-25","paper":"/paper/learning-to-classify-images-without-labels","paper_url":"https://arxiv.org/abs/2005.12320v2","paper_title":"SCAN: Learning to Classify Images without Labels","code":"https://github.com/wvangansbeke/Unsupervised-Classification","n_code_links":2,"syntology":null},{"rank_in_archive_order":54,"model":"OBoW (ResNet-50)","metrics":{"Top 5 Accuracy":"82.9%"},"uses_additional_data":false,"paper_date":"2020-12-21","paper":"/paper/online-bag-of-visual-words-generation-for","paper_url":"https://arxiv.org/abs/2012.11552v2","paper_title":"OBoW: Online Bag-of-Visual-Words Generation for Self-Supervised Learning","code":"https://github.com/valeoai/obow","n_code_links":3,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":55,"model":"PCL (ResNet-50)","metrics":{"Top 5 Accuracy":"75.6%"},"uses_additional_data":false,"paper_date":"2020-05-11","paper":"/paper/prototypical-contrastive-learning-of","paper_url":"https://arxiv.org/abs/2005.04966v5","paper_title":"Prototypical Contrastive Learning of Unsupervised Representations","code":"https://github.com/salesforce/PCL","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":1}},{"rank_in_archive_order":56,"model":"CPC","metrics":{"Top 5 Accuracy":"64.03%"},"uses_additional_data":false,"paper_date":"2018-07-10","paper":"/paper/representation-learning-with-contrastive","paper_url":"http://arxiv.org/abs/1807.03748v2","paper_title":"Representation Learning with Contrastive Predictive Coding","code":"https://github.com/RElbers/info-nce-pytorch","n_code_links":28,"syntology":{"n_ran":29,"n_unverified":16,"n_samples":45,"n_pointer_only_licence":22}},{"rank_in_archive_order":57,"model":"BigBiGAN (RevNet-50 ×4, BN+CReLU)","metrics":{"Top 5 Accuracy":"55.2%"},"uses_additional_data":false,"paper_date":"2019-07-04","paper":"/paper/large-scale-adversarial-representation","paper_url":"https://arxiv.org/abs/1907.02544v2","paper_title":"Large Scale Adversarial Representation Learning","code":"https://github.com/lukemelas/unsupervised-image-segmentation","n_code_links":4,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":58,"model":"Rotation (joint training)","metrics":{"Top 5 Accuracy":"53.37%"},"uses_additional_data":false,"paper_date":"2019-05-09","paper":"/paper/190503670","paper_url":"https://arxiv.org/abs/1905.03670v2","paper_title":"S4L: Self-Supervised Semi-Supervised Learning","code":"https://github.com/google-research/s4l","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":19,"n_samples":19,"n_pointer_only_licence":19}},{"rank_in_archive_order":59,"model":"Pseudolabeling","metrics":{"Top 5 Accuracy":"51.56%"},"uses_additional_data":false,"paper_date":"2019-05-09","paper":"/paper/190503670","paper_url":"https://arxiv.org/abs/1905.03670v2","paper_title":"S4L: Self-Supervised Semi-Supervised Learning","code":"https://github.com/google-research/s4l","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":19,"n_samples":19,"n_pointer_only_licence":19}},{"rank_in_archive_order":60,"model":"Exemplar (joint training)","metrics":{"Top 5 Accuracy":"47.02%"},"uses_additional_data":false,"paper_date":"2019-05-09","paper":"/paper/190503670","paper_url":"https://arxiv.org/abs/1905.03670v2","paper_title":"S4L: Self-Supervised Semi-Supervised Learning","code":"https://github.com/google-research/s4l","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":19,"n_samples":19,"n_pointer_only_licence":19}},{"rank_in_archive_order":61,"model":"VAT + Entropy Minimization","metrics":{"Top 5 Accuracy":"46.96%"},"uses_additional_data":false,"paper_date":"2019-05-09","paper":"/paper/190503670","paper_url":"https://arxiv.org/abs/1905.03670v2","paper_title":"S4L: Self-Supervised Semi-Supervised Learning","code":"https://github.com/google-research/s4l","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":19,"n_samples":19,"n_pointer_only_licence":19}},{"rank_in_archive_order":62,"model":"Rotation","metrics":{"Top 5 Accuracy":"45.11%"},"uses_additional_data":false,"paper_date":"2019-05-09","paper":"/paper/190503670","paper_url":"https://arxiv.org/abs/1905.03670v2","paper_title":"S4L: Self-Supervised Semi-Supervised Learning","code":"https://github.com/google-research/s4l","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":19,"n_samples":19,"n_pointer_only_licence":19}},{"rank_in_archive_order":63,"model":"Exemplar","metrics":{"Top 5 Accuracy":"44.90%"},"uses_additional_data":false,"paper_date":"2019-05-09","paper":"/paper/190503670","paper_url":"https://arxiv.org/abs/1905.03670v2","paper_title":"S4L: Self-Supervised Semi-Supervised Learning","code":"https://github.com/google-research/s4l","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":19,"n_samples":19,"n_pointer_only_licence":19}},{"rank_in_archive_order":64,"model":"VAT","metrics":{"Top 5 Accuracy":"44.05%"},"uses_additional_data":false,"paper_date":"2019-05-09","paper":"/paper/190503670","paper_url":"https://arxiv.org/abs/1905.03670v2","paper_title":"S4L: Self-Supervised Semi-Supervised Learning","code":"https://github.com/google-research/s4l","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":19,"n_samples":19,"n_pointer_only_licence":19}},{"rank_in_archive_order":65,"model":"Instance Discrimination (ResNet-50)","metrics":{"Top 5 Accuracy":"39.20%"},"uses_additional_data":false,"paper_date":"2018-06-01","paper":"/paper/unsupervised-feature-learning-via-non-1","paper_url":"http://openaccess.thecvf.com/content_cvpr_2018/html/Wu_Unsupervised_Feature_Learning_CVPR_2018_paper.html","paper_title":"Unsupervised Feature Learning via Non-Parametric Instance Discrimination","code":"https://github.com/open-mmlab/mmselfsup","n_code_links":4,"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":41,"rows_with_any_sample_ran":24,"distinct_papers_with_graph_line":23,"distinct_papers_with_any_sample_ran":17,"samples_over_distinct_papers":{"n_ran":254,"n_unverified":193,"n_samples":447,"n_pointer_only_licence":168,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":612,"n_unverified":532,"n_samples":1144,"n_pointer_only_licence":530,"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"}}}