{"url":"/sota/zero-shot-transfer-image-classification-on-1","task":{"name":"Zero-Shot Transfer Image Classification","url":"/task/zero-shot-transfer-image-classification","note":null},"dataset":{"name":"ImageNet","url":"/dataset/imagenet"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":null,"description_from":null,"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":["Param","Accuracy (Private)","Accuracy (Public)"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Param":null,"Accuracy (Private)":"higher","Accuracy (Public)":"higher"}},"counts":{"rows":23,"rows_with_code":18,"rows_with_paper_page":20,"rows_dated":20,"rows_using_additional_data":6},"rows":[{"rank_in_archive_order":1,"model":"M2-Encoder","metrics":{"Accuracy (Private)":"88.5","Param":"10B"},"uses_additional_data":true,"paper_date":"2024-01-29","paper":"/paper/boldsymbol-m-2-encoder-advancing-bilingual","paper_url":"https://arxiv.org/abs/2401.15896v2","paper_title":"M2-Encoder: Advancing Bilingual Image-Text Understanding by Large-scale Efficient Pretraining","code":"https://github.com/alipay/Ant-Multi-Modal-Framework/tree/main/prj/M2_Encoder","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"BASIC (Lion)","metrics":{"Accuracy (Private)":"88.3"},"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":3,"model":"CoCa","metrics":{"Accuracy (Private)":"86.3"},"uses_additional_data":true,"paper_date":"2022-05-04","paper":"/paper/coca-contrastive-captioners-are-image-text","paper_url":"https://arxiv.org/abs/2205.01917v2","paper_title":"CoCa: Contrastive Captioners are Image-Text Foundation Models","code":"https://github.com/mlfoundations/open_clip","n_code_links":6,"syntology":{"n_ran":9,"n_unverified":8,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"LiT-22B","metrics":{"Accuracy (Private)":"85.9"},"uses_additional_data":false,"paper_date":"2023-02-10","paper":"/paper/scaling-vision-transformers-to-22-billion","paper_url":"https://arxiv.org/abs/2302.05442v1","paper_title":"Scaling Vision Transformers to 22 Billion Parameters","code":"https://github.com/lucidrains/flash-cosine-sim-attention","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"BASIC","metrics":{"Accuracy (Private)":"85.7"},"uses_additional_data":true,"paper_date":"2021-11-19","paper":"/paper/combined-scaling-for-zero-shot-transfer","paper_url":"https://arxiv.org/abs/2111.10050v3","paper_title":"Combined Scaling for Zero-shot Transfer Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"LiT ViT-e","metrics":{"Accuracy (Private)":"85.4"},"uses_additional_data":false,"paper_date":"2022-09-14","paper":"/paper/pali-a-jointly-scaled-multilingual-language","paper_url":"https://arxiv.org/abs/2209.06794v4","paper_title":"PaLI: A Jointly-Scaled Multilingual Language-Image Model","code":"https://github.com/google-research/big_vision","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"LiT-tuning","metrics":{"Accuracy (Private)":"84.5","Accuracy (Public)":"75.7"},"uses_additional_data":false,"paper_date":"2021-11-15","paper":"/paper/lit-zero-shot-transfer-with-locked-image-text","paper_url":"https://arxiv.org/abs/2111.07991v3","paper_title":"LiT: Zero-Shot Transfer with Locked-image text Tuning","code":"https://github.com/mlfoundations/open_clip","n_code_links":5,"syntology":null},{"rank_in_archive_order":8,"model":"IMP-MoE-L","metrics":{"Accuracy (Private)":"83.9"},"uses_additional_data":false,"paper_date":"2023-05-10","paper":"/paper/alternating-gradient-descent-and-mixture-of","paper_url":"https://arxiv.org/abs/2305.06324v2","paper_title":"Alternating Gradient Descent and Mixture-of-Experts for Integrated Multimodal Perception","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"EVA-CLIP-18B","metrics":{"Accuracy (Private)":"83.8"},"uses_additional_data":false,"paper_date":"2024-02-06","paper":"/paper/eva-clip-18b-scaling-clip-to-18-billion","paper_url":"https://arxiv.org/abs/2402.04252v1","paper_title":"EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters","code":"https://github.com/baaivision/EVA/tree/master/EVA-CLIP-18B","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":5,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"InternVL-C","metrics":{"Accuracy (Private)":"83.2"},"uses_additional_data":false,"paper_date":"2023-12-21","paper":"/paper/internvl-scaling-up-vision-foundation-models","paper_url":"https://arxiv.org/abs/2312.14238v3","paper_title":"InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks","code":"https://github.com/opengvlab/internvl","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":11,"model":"MAWS (ViT-2B)","metrics":{"Accuracy (Private)":"82.1"},"uses_additional_data":false,"paper_date":"2023-03-23","paper":"/paper/the-effectiveness-of-mae-pre-pretraining-for","paper_url":"https://arxiv.org/abs/2303.13496v3","paper_title":"The effectiveness of MAE pre-pretraining for billion-scale pretraining","code":"https://github.com/facebookresearch/maws","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"EVA-CLIP-E/14+","metrics":{"Accuracy (Private)":"82"},"uses_additional_data":false,"paper_date":"2023-03-27","paper":"/paper/eva-clip-improved-training-techniques-for","paper_url":"https://arxiv.org/abs/2303.15389v1","paper_title":"EVA-CLIP: Improved Training Techniques for CLIP at Scale","code":"https://github.com/baaivision/eva","n_code_links":4,"syntology":{"n_ran":0,"n_unverified":4,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"CLIPA (ViT-H/14-336px)","metrics":{"Accuracy (Private)":"81.8"},"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":14,"model":"MAWS (ViT-H)","metrics":{"Accuracy (Private)":"81.1"},"uses_additional_data":false,"paper_date":"2023-03-23","paper":"/paper/the-effectiveness-of-mae-pre-pretraining-for","paper_url":"https://arxiv.org/abs/2303.13496v3","paper_title":"The effectiveness of MAE pre-pretraining for billion-scale pretraining","code":"https://github.com/facebookresearch/maws","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"REACT","metrics":{"Accuracy (Private)":"78.5"},"uses_additional_data":false,"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":16,"model":"ALIGN","metrics":{"Accuracy (Private)":"76.4","Accuracy (Public)":"-"},"uses_additional_data":false,"paper_date":"2021-02-11","paper":"/paper/scaling-up-visual-and-vision-language","paper_url":"https://arxiv.org/abs/2102.05918v2","paper_title":"Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision","code":"https://github.com/facebookresearch/metaclip","n_code_links":5,"syntology":{"n_ran":8,"n_unverified":2,"n_samples":10,"n_pointer_only_licence":9}},{"rank_in_archive_order":17,"model":"CLIP（ViT-L/14-336px）","metrics":{"Accuracy (Private)":"76.2"},"uses_additional_data":true,"paper_date":"2021-02-26","paper":"/paper/learning-transferable-visual-models-from","paper_url":"https://arxiv.org/abs/2103.00020v1","paper_title":"Learning Transferable Visual Models From Natural Language Supervision","code":"https://github.com/openai/CLIP","n_code_links":82,"syntology":{"n_ran":16,"n_unverified":4,"n_samples":20,"n_pointer_only_licence":16}},{"rank_in_archive_order":18,"model":"AltCLIP","metrics":{"Accuracy (Private)":"74.5"},"uses_additional_data":false,"paper_date":"2022-11-12","paper":"/paper/altclip-altering-the-language-encoder-in-clip","paper_url":"https://arxiv.org/abs/2211.06679v2","paper_title":"AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities","code":"https://github.com/flagai-open/flagai","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":9,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"PaLI","metrics":{"Accuracy (Private)":"72.11"},"uses_additional_data":true,"paper_date":"2022-09-14","paper":"/paper/pali-a-jointly-scaled-multilingual-language","paper_url":"https://arxiv.org/abs/2209.06794v4","paper_title":"PaLI: A Jointly-Scaled Multilingual Language-Image Model","code":"https://github.com/google-research/big_vision","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":20,"model":"Diffusion Classifier (zero-shot)","metrics":{"Accuracy (Private)":"61.4"},"uses_additional_data":false,"paper_date":"2023-03-28","paper":"/paper/your-diffusion-model-is-secretly-a-zero-shot","paper_url":"https://arxiv.org/abs/2303.16203v3","paper_title":"Your Diffusion Model is Secretly a Zero-Shot Classifier","code":"https://github.com/diffusion-classifier/diffusion-classifier","n_code_links":4,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":1}},{"rank_in_archive_order":21,"model":"CLIP (ResNet50)","metrics":{"Accuracy (Private)":"59.6"},"uses_additional_data":true,"paper_date":"2021-02-26","paper":"/paper/learning-transferable-visual-models-from","paper_url":"https://arxiv.org/abs/2103.00020v1","paper_title":"Learning Transferable Visual Models From Natural Language Supervision","code":"https://github.com/openai/CLIP","n_code_links":82,"syntology":{"n_ran":16,"n_unverified":4,"n_samples":20,"n_pointer_only_licence":16}},{"rank_in_archive_order":22,"model":"CWCL","metrics":{"Accuracy (Public)":"76.5"},"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":23,"model":"CLIP","metrics":{"Accuracy (Public)":" 31.3"},"uses_additional_data":false,"paper_date":"2021-02-26","paper":"/paper/learning-transferable-visual-models-from","paper_url":"https://arxiv.org/abs/2103.00020v1","paper_title":"Learning Transferable Visual Models From Natural Language Supervision","code":"https://github.com/openai/CLIP","n_code_links":82,"syntology":{"n_ran":16,"n_unverified":4,"n_samples":20,"n_pointer_only_licence":16}}],"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":13,"rows_with_any_sample_ran":12,"distinct_papers_with_graph_line":10,"distinct_papers_with_any_sample_ran":9,"samples_over_distinct_papers":{"n_ran":48,"n_unverified":45,"n_samples":93,"n_pointer_only_licence":28,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":82,"n_unverified":55,"n_samples":137,"n_pointer_only_licence":60,"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"}}}