{"url":"/sota/zero-shot-video-retrieval-on-lsmdc","task":{"name":"Zero-Shot Video Retrieval","url":"/task/zero-shot-video-retrieval","note":null},"dataset":{"name":"LSMDC","url":"/dataset/lsmdc"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"Zero-shot video retrieval is the task of retrieving relevant videos based on a query (usually in text form) without any prior training on specific examples of those videos. Unlike traditional retrieval methods that rely on supervised learning with annotated datasets, zero-shot retrieval leverages pre-trained models, typically based on large-scale vision-language learning, to understand semantic relationships between textual descriptions and video content.\r\n\r\nThis approach enables retrieval of unseen video concepts by generalizing knowledge from diverse training data, making it highly useful for domains with limited labeled data, such as broadcast media, surveillance, and historical archives.","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":["text-to-video R@1","text-to-video R@5","text-to-video R@10","text-to-video Median Rank","text-to-video Mean Rank","video-to-text R@1","video-to-text R@5","video-to-text R@10"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"text-to-video R@1":null,"text-to-video R@5":null,"text-to-video R@10":null,"text-to-video Median Rank":null,"text-to-video Mean Rank":null,"video-to-text R@1":null,"video-to-text R@5":null,"video-to-text R@10":null}},"counts":{"rows":16,"rows_with_code":14,"rows_with_paper_page":16,"rows_dated":16,"rows_using_additional_data":8},"rows":[{"rank_in_archive_order":1,"model":"InternVideo2-6B","metrics":{"text-to-video R@1":"33.8","text-to-video R@10":"62.2","text-to-video R@5":"55.9","video-to-text R@1":"30.1","video-to-text R@10":"54.8","video-to-text R@5":"47.7"},"uses_additional_data":true,"paper_date":"2024-03-22","paper":"/paper/internvideo2-scaling-video-foundation-models","paper_url":"https://arxiv.org/abs/2403.15377v4","paper_title":"InternVideo2: Scaling Foundation Models for Multimodal Video Understanding","code":"https://github.com/opengvlab/internvideo","n_code_links":2,"syntology":null},{"rank_in_archive_order":2,"model":"InternVideo2-1B","metrics":{"text-to-video R@1":"32.0","text-to-video R@10":"59.4","text-to-video R@5":"52.4","video-to-text R@1":"27.3","video-to-text R@10":"51.6","video-to-text R@5":"44.2"},"uses_additional_data":true,"paper_date":"2024-03-22","paper":"/paper/internvideo2-scaling-video-foundation-models","paper_url":"https://arxiv.org/abs/2403.15377v4","paper_title":"InternVideo2: Scaling Foundation Models for Multimodal Video Understanding","code":"https://github.com/opengvlab/internvideo","n_code_links":2,"syntology":null},{"rank_in_archive_order":3,"model":"VAST, HowToCaption-finetuned","metrics":{"text-to-video Median Rank":"7","text-to-video R@1":"27.7","text-to-video R@10":"54.6","text-to-video R@5":"46.5"},"uses_additional_data":false,"paper_date":"2023-10-07","paper":"/paper/howtocaption-prompting-llms-to-transform","paper_url":"https://arxiv.org/abs/2310.04900v2","paper_title":"HowToCaption: Prompting LLMs to Transform Video Annotations at Scale","code":"https://github.com/ninatu/howtocaption","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":1,"n_samples":5,"n_pointer_only_licence":5}},{"rank_in_archive_order":4,"model":"UMT-L (ViT-L/16)","metrics":{"text-to-video R@1":"25.2","text-to-video R@10":"50.5","text-to-video R@5":"43.0","video-to-text R@1":"23.2","video-to-text R@10":" 44.2","video-to-text R@5":"37.7"},"uses_additional_data":true,"paper_date":"2023-03-28","paper":"/paper/unmasked-teacher-towards-training-efficient","paper_url":"https://arxiv.org/abs/2303.16058v2","paper_title":"Unmasked Teacher: Towards Training-Efficient Video Foundation Models","code":"https://github.com/opengvlab/unmasked_teacher","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":5,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"mPLUG-2","metrics":{"text-to-video R@1":"24.1","text-to-video R@10":"52.0","text-to-video R@5":"43.8"},"uses_additional_data":false,"paper_date":"2023-02-01","paper":"/paper/mplug-2-a-modularized-multi-modal-foundation","paper_url":"https://arxiv.org/abs/2302.00402v1","paper_title":"mPLUG-2: A Modularized Multi-modal Foundation Model Across Text, Image and Video","code":"https://github.com/modelscope/modelscope","n_code_links":4,"syntology":{"n_ran":9,"n_unverified":10,"n_samples":19,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"BT-Adapter","metrics":{"text-to-video R@1":"19.5","text-to-video R@10":"45.0","text-to-video R@5":"35.9"},"uses_additional_data":false,"paper_date":"2023-09-27","paper":"/paper/one-for-all-video-conversation-is-feasible","paper_url":"https://arxiv.org/abs/2309.15785v2","paper_title":"BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning","code":"https://github.com/farewellthree/BT-Adapter","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":3,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"HiTeA-17M","metrics":{"text-to-video R@1":"18.3","text-to-video R@10":"44.2","text-to-video R@5":"36.7"},"uses_additional_data":true,"paper_date":"2022-12-30","paper":"/paper/hitea-hierarchical-temporal-aware-video","paper_url":"https://arxiv.org/abs/2212.14546v1","paper_title":"HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"InternVideo","metrics":{"text-to-video R@1":"17.6","text-to-video R@10":"40.2","text-to-video R@5":"32.4","video-to-text R@1":"13.2","video-to-text R@10":"34.9","video-to-text R@5":"27.8"},"uses_additional_data":true,"paper_date":"2022-12-06","paper":"/paper/internvideo-general-video-foundation-models","paper_url":"https://arxiv.org/abs/2212.03191v2","paper_title":"InternVideo: General Video Foundation Models via Generative and Discriminative Learning","code":"https://github.com/opengvlab/internvideo","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"HowToCaption","metrics":{"text-to-video Median Rank":"29","text-to-video R@1":"17.3","text-to-video R@10":"38.6","text-to-video R@5":"31.7"},"uses_additional_data":false,"paper_date":"2023-10-07","paper":"/paper/howtocaption-prompting-llms-to-transform","paper_url":"https://arxiv.org/abs/2310.04900v2","paper_title":"HowToCaption: Prompting LLMs to Transform Video Annotations at Scale","code":"https://github.com/ninatu/howtocaption","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":1,"n_samples":5,"n_pointer_only_licence":5}},{"rank_in_archive_order":10,"model":"Yatai Ji et. al.","metrics":{"text-to-video R@1":"17.2","text-to-video R@10":"39.1","text-to-video R@5":"32.4"},"uses_additional_data":false,"paper_date":"2022-11-24","paper":"/paper/seeing-what-you-miss-vision-language-pre","paper_url":"https://arxiv.org/abs/2211.13437v2","paper_title":"Seeing What You Miss: Vision-Language Pre-training with Semantic Completion Learning","code":"https://github.com/iigroup/scl","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"HiTeA-5M","metrics":{"text-to-video R@1":"15.5","text-to-video R@10":"39.8","text-to-video R@5":"31.1"},"uses_additional_data":true,"paper_date":"2022-12-30","paper":"/paper/hitea-hierarchical-temporal-aware-video","paper_url":"https://arxiv.org/abs/2212.14546v1","paper_title":"HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"CLIP4Clip","metrics":{"text-to-video Mean Rank":"117","text-to-video Median Rank":"28","text-to-video R@1":"15.1","text-to-video R@10":"36.4","text-to-video R@5":"28.5"},"uses_additional_data":true,"paper_date":"2021-04-18","paper":"/paper/clip4clip-an-empirical-study-of-clip-for-end","paper_url":"https://arxiv.org/abs/2104.08860v2","paper_title":"CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval","code":"https://github.com/towhee-io/towhee","n_code_links":5,"syntology":{"n_ran":3,"n_unverified":1,"n_samples":4,"n_pointer_only_licence":3}},{"rank_in_archive_order":13,"model":"Clover","metrics":{"text-to-video Median Rank":"24","text-to-video R@1":"14.7","text-to-video R@10":"38.2","text-to-video R@5":"29.2"},"uses_additional_data":true,"paper_date":"2022-07-16","paper":"/paper/clover-towards-a-unified-video-language","paper_url":"https://arxiv.org/abs/2207.07885v3","paper_title":"Clover: Towards A Unified Video-Language Alignment and Fusion Model","code":"https://github.com/leeyn-43/clover","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"Y. Ge et. al.","metrics":{"text-to-video Median Rank":" 42.0","text-to-video R@1":"12.2","text-to-video R@10":"32.2","text-to-video R@5":"25.9"},"uses_additional_data":false,"paper_date":"2022-01-13","paper":"/paper/bridgeformer-bridging-video-text-retrieval","paper_url":"https://arxiv.org/abs/2201.04850v2","paper_title":"Bridging Video-text Retrieval with Multiple Choice Questions","code":"https://github.com/towhee-io/towhee","n_code_links":2,"syntology":{"n_ran":13,"n_unverified":11,"n_samples":24,"n_pointer_only_licence":6}},{"rank_in_archive_order":15,"model":"MILES","metrics":{"text-to-video Median Rank":"50.7","text-to-video R@1":"11.1","text-to-video R@10":"30.6","text-to-video R@5":"24.7"},"uses_additional_data":false,"paper_date":"2022-04-26","paper":"/paper/miles-visual-bert-pre-training-with-injected","paper_url":"https://arxiv.org/abs/2204.12408v1","paper_title":"MILES: Visual BERT Pre-training with Injected Language Semantics for Video-text Retrieval","code":"https://github.com/tencentarc/mcq","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"SSML","metrics":{"text-to-video R@1":"4.2","text-to-video R@10":"17.1","text-to-video R@5":"11.6"},"uses_additional_data":false,"paper_date":"2020-03-06","paper":"/paper/noise-estimation-using-density-estimation-for","paper_url":"https://arxiv.org/abs/2003.03186v3","paper_title":"Noise Estimation Using Density Estimation for Self-Supervised Multimodal Learning","code":"https://github.com/elad-amrani/ssml","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 6,795 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":6795,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2785},"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":8,"rows_with_any_sample_ran":8,"distinct_papers_with_graph_line":7,"distinct_papers_with_any_sample_ran":7,"samples_over_distinct_papers":{"n_ran":36,"n_unverified":31,"n_samples":67,"n_pointer_only_licence":14,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":40,"n_unverified":32,"n_samples":72,"n_pointer_only_licence":19,"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"}}}