{"about":{"non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","site":"https://codewithpapers.app","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/dataset/msr-vtt/papers/ran/1","list_of":"/dataset/msr-vtt","dataset":"MSR-VTT","archive":{"snapshot":"2025-07-28"},"syntology_read_at":"2026-09-28T10:30:06+00:00","key_notes":{"samples_ran_checked":"legacy name, kept unchanged so existing readers do not break: it counts the samples that ran with no instrument failure (honoured, violated, and ran with no contract checked); it does not mean a contract was checked, and the pages print it as 'K with no instrument failure', not 'K checked'","samples_constructed":"a sub-count of the samples that ran, never subtracted from them and never a failure: an executed sample whose run returned an instance of its own class (fixture_out_type equals the entry name): the run built an object and did not compute a result (Syntology's RAN record, counts.constructed)"},"order":"ran","order_definition":"only papers where Syntology ran at least one harvested sample; date (newest first), ties by arXiv id","caption":"We ran code from the paper's repository; we did not run it on this dataset or check it against this dataset's benchmarks.","absence":"A paper missing from this list is not a recorded non-run: it may have no arXiv id, no harvested code, or only samples that have not run yet.","population":"every paper with a leaderboard row on this dataset's benchmarks (the benchmark-backed subset): the archive's own papers-using-this-dataset list was never published, so this is not that list; num_papers_in_archive is the archive's own count","page":1,"pages_in_order":1,"rows_per_page":100,"rows":[1,49],"of":49,"counts":{"papers_with_a_benchmark_row":117,"with_a_code_link":92,"where_syntology_ran_a_sample":49,"not_listed_spam_title":0,"listed":117,"listed_where_code_ran":49,"where_syntology_ran_a_sample_split":{"with_a_run_with_no_instrument_failure":42,"every_run_a_failure_of_syntologys_instrument":7,"listed_with_a_run_with_no_instrument_failure":42,"listed_every_run_a_failure_of_syntologys_instrument":7,"filter":{"states":["a run with no instrument failure","any run, instrument failures included"],"default":"a run with no instrument failure","note":"on the 'only where code ran' pages the default hides, in the browser, the rows where every run was a failure of Syntology's instrument; the second state shows them again. Rows are hidden, never re-ordered; these twins list every row"}},"definition":"distinct papers with at least one leaderboard row on this dataset's benchmarks; 'where Syntology ran a sample' counts papers with at least one harvested sample that ran, which is not a correctness claim"},"first_page":"/dataset/msr-vtt/papers/ran/1","prev":null,"next":null,"papers":[{"paper":"/paper/make-your-training-flexible-towards","slug":"make-your-training-flexible-towards","title":"Make Your Training Flexible: Towards Deployment-Efficient Video Models","date":"2025-03-18","arxiv_id":"2503.14237","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":17,"samples_ran":17,"samples_constructed":0,"samples_ran_checked":17,"samples_ran_instrument_failed":0,"samples_unverified":0,"pointer_only_for_licence":6,"official":{"repos":["opengvlab/fluxvit"],"state":"official (archive's flag): 17 ran","n_ran":17,"n_constructed":0,"n_ran_no_instrument_failure":17,"n_unverified":0,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/make-your-training-flexible-towards#ran","syntology_url":"https://syntology.ai/paper/2503.14237","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.14237"}}}},{"paper":"/paper/gramian-multimodal-representation-learning","slug":"gramian-multimodal-representation-learning","title":"Gramian Multimodal Representation Learning and Alignment","date":"2024-12-16","arxiv_id":"2412.11959","rows_on_this_dataset":2,"code_links":2,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":12,"samples_ran":7,"samples_constructed":0,"samples_ran_checked":7,"samples_ran_instrument_failed":0,"samples_unverified":5,"pointer_only_for_licence":1,"official":{"repos":["ispamm/GRAM"],"state":"official (archive's flag): 7 ran","n_ran":7,"n_constructed":0,"n_ran_no_instrument_failure":7,"n_unverified":5,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/gramian-multimodal-representation-learning#ran","syntology_url":"https://syntology.ai/paper/2412.11959","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2412.11959"}}}},{"paper":"/paper/vid-tldr-training-free-token-merging-for","slug":"vid-tldr-training-free-token-merging-for","title":"vid-TLDR: Training Free Token merging for Light-weight Video Transformer","date":"2024-03-20","arxiv_id":"2403.13347","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":4,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":0,"samples_ran_instrument_failed":3,"samples_unverified":1,"pointer_only_for_licence":0,"official":{"repos":["mlvlab/vid-tldr"],"state":"official (archive's flag): 3 ran","n_ran":3,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":1,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/vid-tldr-training-free-token-merging-for#ran","syntology_url":"https://syntology.ai/paper/2403.13347","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.13347"}}}},{"paper":"/paper/video-lavit-unified-video-language-pre","slug":"video-lavit-unified-video-language-pre","title":"Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization","date":"2024-02-05","arxiv_id":"2402.03161","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":5,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":1,"samples_ran_instrument_failed":2,"samples_unverified":2,"pointer_only_for_licence":5,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/video-lavit-unified-video-language-pre#ran","syntology_url":"https://syntology.ai/paper/2402.03161","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2402.03161"}}}},{"paper":"/paper/multi-granularity-correspondence-learning-1","slug":"multi-granularity-correspondence-learning-1","title":"Multi-granularity Correspondence Learning from Long-term Noisy Videos","date":"2024-01-30","arxiv_id":"2401.16702","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":3,"samples_ran":2,"samples_constructed":2,"samples_ran_checked":2,"samples_ran_instrument_failed":0,"samples_unverified":1,"pointer_only_for_licence":0,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/multi-granularity-correspondence-learning-1#ran","syntology_url":"https://syntology.ai/paper/2401.16702","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2401.16702"}}}},{"paper":"/paper/videoassembler-identity-consistent-video","slug":"videoassembler-identity-consistent-video","title":"MagDiff: Multi-Alignment Diffusion for High-Fidelity Video Generation and Editing","date":"2023-11-29","arxiv_id":"2311.17338","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":12,"samples_ran":10,"samples_constructed":0,"samples_ran_checked":9,"samples_ran_instrument_failed":1,"samples_unverified":2,"pointer_only_for_licence":12,"official":{"repos":["gulucaptain/videoassembler"],"state":"official (archive's flag): 10 ran","n_ran":10,"n_constructed":0,"n_ran_no_instrument_failure":9,"n_unverified":2,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/videoassembler-identity-consistent-video#ran","syntology_url":"https://syntology.ai/paper/2311.17338","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2311.17338"}}}},{"paper":"/paper/howtocaption-prompting-llms-to-transform","slug":"howtocaption-prompting-llms-to-transform","title":"HowToCaption: Prompting LLMs to Transform Video Annotations at Scale","date":"2023-10-07","arxiv_id":"2310.04900","rows_on_this_dataset":3,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":5,"samples_ran":4,"samples_constructed":0,"samples_ran_checked":4,"samples_ran_instrument_failed":0,"samples_unverified":1,"pointer_only_for_licence":5,"official":{"repos":["ninatu/howtocaption"],"state":"official (archive's flag): 4 ran","n_ran":4,"n_constructed":0,"n_ran_no_instrument_failure":4,"n_unverified":1,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/howtocaption-prompting-llms-to-transform#ran","syntology_url":"https://syntology.ai/paper/2310.04900","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.04900"}}}},{"paper":"/paper/languagebind-extending-video-language","slug":"languagebind-extending-video-language","title":"LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment","date":"2023-10-03","arxiv_id":"2310.01852","rows_on_this_dataset":2,"code_links":6,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":14,"samples_ran":10,"samples_constructed":0,"samples_ran_checked":9,"samples_ran_instrument_failed":1,"samples_unverified":4,"pointer_only_for_licence":3,"official":{"repos":["pku-yuangroup/languagebind"],"state":"official (archive's flag): 5 ran","n_ran":5,"n_constructed":0,"n_ran_no_instrument_failure":5,"n_unverified":0,"ran_from_kinds":["listed","official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/languagebind-extending-video-language#ran","syntology_url":"https://syntology.ai/paper/2310.01852","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.01852"}}}},{"paper":"/paper/prototype-based-aleatoric-uncertainty-1","slug":"prototype-based-aleatoric-uncertainty-1","title":"Prototype-based Aleatoric Uncertainty Quantification for Cross-modal Retrieval","date":"2023-09-29","arxiv_id":"2309.17093","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":19,"samples_ran":17,"samples_constructed":0,"samples_ran_checked":10,"samples_ran_instrument_failed":7,"samples_unverified":2,"pointer_only_for_licence":8,"official":{"repos":["leolee99/pau"],"state":"official (archive's flag): 17 ran","n_ran":17,"n_constructed":0,"n_ran_no_instrument_failure":10,"n_unverified":2,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/prototype-based-aleatoric-uncertainty-1#ran","syntology_url":"https://syntology.ai/paper/2309.17093","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2309.17093"}}}},{"paper":"/paper/one-for-all-video-conversation-is-feasible","slug":"one-for-all-video-conversation-is-feasible","title":"BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning","date":"2023-09-27","arxiv_id":"2309.15785","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":4,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":2,"samples_ran_instrument_failed":1,"samples_unverified":1,"pointer_only_for_licence":2,"official":{"repos":["farewellthree/BT-Adapter"],"state":"official (archive's flag): 3 ran","n_ran":3,"n_constructed":0,"n_ran_no_instrument_failure":2,"n_unverified":1,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/one-for-all-video-conversation-is-feasible#ran","syntology_url":"https://syntology.ai/paper/2309.15785","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2309.15785"}}}},{"paper":"/paper/show-1-marrying-pixel-and-latent-diffusion","slug":"show-1-marrying-pixel-and-latent-diffusion","title":"Show-1: Marrying Pixel and Latent Diffusion Models for Text-to-Video Generation","date":"2023-09-27","arxiv_id":"2309.15818","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":1,"samples_ran":1,"samples_constructed":0,"samples_ran_checked":1,"samples_ran_instrument_failed":0,"samples_unverified":0,"pointer_only_for_licence":1,"official":{"repos":["showlab/show-1"],"state":"official (archive's flag): 1 ran","n_ran":1,"n_constructed":0,"n_ran_no_instrument_failure":1,"n_unverified":0,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/show-1-marrying-pixel-and-latent-diffusion#ran","syntology_url":"https://syntology.ai/paper/2309.15818","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2309.15818"}}}},{"paper":"/paper/accurate-and-fast-compressed-video-captioning","slug":"accurate-and-fast-compressed-video-captioning","title":"Accurate and Fast Compressed Video Captioning","date":"2023-09-22","arxiv_id":"2309.12867","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":14,"samples_ran":10,"samples_constructed":8,"samples_ran_checked":9,"samples_ran_instrument_failed":1,"samples_unverified":4,"pointer_only_for_licence":0,"official":{"repos":["acherstyx/CoCap"],"state":"official (archive's flag): 10 ran","n_ran":10,"n_constructed":8,"n_ran_no_instrument_failure":9,"n_unverified":4,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/accurate-and-fast-compressed-video-captioning#ran","syntology_url":"https://syntology.ai/paper/2309.12867","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2309.12867"}}}},{"paper":"/paper/unified-coarse-to-fine-alignment-for-video","slug":"unified-coarse-to-fine-alignment-for-video","title":"Unified Coarse-to-Fine Alignment for Video-Text Retrieval","date":"2023-09-18","arxiv_id":"2309.10091","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":14,"samples_ran":11,"samples_constructed":0,"samples_ran_checked":8,"samples_ran_instrument_failed":3,"samples_unverified":3,"pointer_only_for_licence":4,"official":{"repos":["ziyang412/ucofia"],"state":"official (archive's flag): 11 ran","n_ran":11,"n_constructed":0,"n_ran_no_instrument_failure":8,"n_unverified":3,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/unified-coarse-to-fine-alignment-for-video#ran","syntology_url":"https://syntology.ai/paper/2309.10091","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2309.10091"}}}},{"paper":"/paper/modelscope-text-to-video-technical-report","slug":"modelscope-text-to-video-technical-report","title":"ModelScope Text-to-Video Technical Report","date":"2023-08-12","arxiv_id":"2308.06571","rows_on_this_dataset":1,"code_links":5,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":16,"samples_ran":11,"samples_constructed":0,"samples_ran_checked":11,"samples_ran_instrument_failed":0,"samples_unverified":5,"pointer_only_for_licence":1,"official":{"repos":["exponentialml/text-to-video-finetuning"],"state":"official (archive's flag): 11 ran","n_ran":11,"n_constructed":0,"n_ran_no_instrument_failure":11,"n_unverified":5,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/modelscope-text-to-video-technical-report#ran","syntology_url":"https://syntology.ai/paper/2308.06571","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2308.06571"}}}},{"paper":"/paper/videocomposer-compositional-video-synthesis","slug":"videocomposer-compositional-video-synthesis","title":"VideoComposer: Compositional Video Synthesis with Motion Controllability","date":"2023-06-03","arxiv_id":"2306.02018","rows_on_this_dataset":1,"code_links":4,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":21,"samples_ran":16,"samples_constructed":0,"samples_ran_checked":12,"samples_ran_instrument_failed":4,"samples_unverified":5,"pointer_only_for_licence":1,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/videocomposer-compositional-video-synthesis#ran","syntology_url":"https://syntology.ai/paper/2306.02018","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2306.02018"}}}},{"paper":"/paper/vast-a-vision-audio-subtitle-text-omni-1","slug":"vast-a-vision-audio-subtitle-text-omni-1","title":"VAST: A Vision-Audio-Subtitle-Text Omni-Modality Foundation Model and Dataset","date":"2023-05-29","arxiv_id":"2305.18500","rows_on_this_dataset":3,"code_links":2,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":42,"samples_ran":35,"samples_constructed":4,"samples_ran_checked":29,"samples_ran_instrument_failed":6,"samples_unverified":7,"pointer_only_for_licence":8,"official":{"repos":["txh-mercury/vast"],"state":"official (archive's flag): 12 ran","n_ran":12,"n_constructed":4,"n_ran_no_instrument_failure":12,"n_unverified":7,"ran_from_kinds":["listed","official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/vast-a-vision-audio-subtitle-text-omni-1#ran","syntology_url":"https://syntology.ai/paper/2305.18500","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.18500"}}}},{"paper":"/paper/imagebind-one-embedding-space-to-bind-them","slug":"imagebind-one-embedding-space-to-bind-them","title":"ImageBind: One Embedding Space To Bind Them All","date":"2023-05-09","arxiv_id":"2305.05665","rows_on_this_dataset":1,"code_links":3,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":34,"samples_ran":24,"samples_constructed":14,"samples_ran_checked":21,"samples_ran_instrument_failed":3,"samples_unverified":10,"pointer_only_for_licence":32,"official":{"repos":["facebookresearch/imagebind"],"state":"official (archive's flag): 18 ran","n_ran":18,"n_constructed":10,"n_ran_no_instrument_failure":15,"n_unverified":7,"ran_from_kinds":["listed","official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/imagebind-one-embedding-space-to-bind-them#ran","syntology_url":"https://syntology.ai/paper/2305.05665","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.05665"}}}},{"paper":"/paper/align-your-latents-high-resolution-video","slug":"align-your-latents-high-resolution-video","title":"Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models","date":"2023-04-18","arxiv_id":"2304.08818","rows_on_this_dataset":2,"code_links":4,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":26,"samples_ran":18,"samples_constructed":0,"samples_ran_checked":12,"samples_ran_instrument_failed":6,"samples_unverified":8,"pointer_only_for_licence":3,"official":{"repos":["stability-ai/generative-models"],"state":"official: no sample here; runs from other or unrecorded repositories","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["found_in_text","listed","unlocated"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/align-your-latents-high-resolution-video#ran","syntology_url":"https://syntology.ai/paper/2304.08818","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2304.08818"}}}},{"paper":"/paper/mammut-a-simple-architecture-for-joint","slug":"mammut-a-simple-architecture-for-joint","title":"MaMMUT: A Simple Architecture for Joint Learning for MultiModal Tasks","date":"2023-03-29","arxiv_id":"2303.16839","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":3,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":3,"samples_ran_instrument_failed":0,"samples_unverified":0,"pointer_only_for_licence":2,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/mammut-a-simple-architecture-for-joint#ran","syntology_url":"https://syntology.ai/paper/2303.16839","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2303.16839"}}}},{"paper":"/paper/unmasked-teacher-towards-training-efficient","slug":"unmasked-teacher-towards-training-efficient","title":"Unmasked Teacher: Towards Training-Efficient Video Foundation Models","date":"2023-03-28","arxiv_id":"2303.16058","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":8,"samples_ran":8,"samples_constructed":0,"samples_ran_checked":7,"samples_ran_instrument_failed":1,"samples_unverified":0,"pointer_only_for_licence":3,"official":{"repos":["opengvlab/unmasked_teacher"],"state":"official (archive's flag): 8 ran","n_ran":8,"n_constructed":0,"n_ran_no_instrument_failure":7,"n_unverified":0,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/unmasked-teacher-towards-training-efficient#ran","syntology_url":"https://syntology.ai/paper/2303.16058","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2303.16058"}}}},{"paper":"/paper/video-text-as-game-players-hierarchical","slug":"video-text-as-game-players-hierarchical","title":"Video-Text as Game Players: Hierarchical Banzhaf Interaction for Cross-Modal Representation Learning","date":"2023-03-25","arxiv_id":"2303.14369","rows_on_this_dataset":1,"code_links":4,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":16,"samples_ran":12,"samples_constructed":7,"samples_ran_checked":11,"samples_ran_instrument_failed":1,"samples_unverified":4,"pointer_only_for_licence":0,"official":{"repos":["jpthu17/HBI"],"state":"official (archive's flag): 1 ran","n_ran":1,"n_constructed":0,"n_ran_no_instrument_failure":1,"n_unverified":0,"ran_from_kinds":["listed","official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/video-text-as-game-players-hierarchical#ran","syntology_url":"https://syntology.ai/paper/2303.14369","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2303.14369"}}}},{"paper":"/paper/meltr-meta-loss-transformer-for-learning-to","slug":"meltr-meta-loss-transformer-for-learning-to","title":"MELTR: Meta Loss Transformer for Learning to Fine-tune Video Foundation Models","date":"2023-03-23","arxiv_id":"2303.13009","rows_on_this_dataset":7,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":1,"samples_ran":1,"samples_constructed":1,"samples_ran_checked":1,"samples_ran_instrument_failed":0,"samples_unverified":0,"pointer_only_for_licence":0,"official":{"repos":["mlvlab/MELTR"],"state":"official (archive's flag): 1 ran","n_ran":1,"n_constructed":1,"n_ran_no_instrument_failure":1,"n_unverified":0,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/meltr-meta-loss-transformer-for-learning-to#ran","syntology_url":"https://syntology.ai/paper/2303.13009","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2303.13009"}}}},{"paper":"/paper/diffusionret-generative-text-video-retrieval","slug":"diffusionret-generative-text-video-retrieval","title":"DiffusionRet: Generative Text-Video Retrieval with Diffusion Model","date":"2023-03-17","arxiv_id":"2303.09867","rows_on_this_dataset":2,"code_links":4,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":6,"samples_ran":5,"samples_constructed":0,"samples_ran_checked":5,"samples_ran_instrument_failed":0,"samples_unverified":1,"pointer_only_for_licence":0,"official":{"repos":["jpthu17/diffusionret"],"state":"official (archive's flag): 5 ran","n_ran":5,"n_constructed":0,"n_ran_no_instrument_failure":5,"n_unverified":1,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/diffusionret-generative-text-video-retrieval#ran","syntology_url":"https://syntology.ai/paper/2303.09867","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2303.09867"}}}},{"paper":"/paper/mplug-2-a-modularized-multi-modal-foundation","slug":"mplug-2-a-modularized-multi-modal-foundation","title":"mPLUG-2: A Modularized Multi-modal Foundation Model Across Text, Image and Video","date":"2023-02-01","arxiv_id":"2302.00402","rows_on_this_dataset":3,"code_links":4,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":19,"samples_ran":17,"samples_constructed":0,"samples_ran_checked":9,"samples_ran_instrument_failed":8,"samples_unverified":2,"pointer_only_for_licence":2,"official":{"repos":["alibaba/AliceMind"],"state":"community repositories only","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["listed"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/mplug-2-a-modularized-multi-modal-foundation#ran","syntology_url":"https://syntology.ai/paper/2302.00402","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2302.00402"}}}},{"paper":"/paper/cap4video-what-can-auxiliary-captions-do-for","slug":"cap4video-what-can-auxiliary-captions-do-for","title":"Cap4Video: What Can Auxiliary Captions Do for Text-Video Retrieval?","date":"2022-12-31","arxiv_id":"2301.00184","rows_on_this_dataset":1,"code_links":4,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":25,"samples_ran":20,"samples_constructed":0,"samples_ran_checked":12,"samples_ran_instrument_failed":8,"samples_unverified":5,"pointer_only_for_licence":11,"official":{"repos":["whwu95/Cap4Video"],"state":"official (archive's flag): 14 ran","n_ran":14,"n_constructed":0,"n_ran_no_instrument_failure":11,"n_unverified":1,"ran_from_kinds":["listed","official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/cap4video-what-can-auxiliary-captions-do-for#ran","syntology_url":"https://syntology.ai/paper/2301.00184","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2301.00184"}}}},{"paper":"/paper/internvideo-general-video-foundation-models","slug":"internvideo-general-video-foundation-models","title":"InternVideo: General Video Foundation Models via Generative and Discriminative Learning","date":"2022-12-06","arxiv_id":"2212.03191","rows_on_this_dataset":2,"code_links":2,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":3,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":0,"samples_ran_instrument_failed":3,"samples_unverified":0,"pointer_only_for_licence":0,"official":{"repos":["opengvlab/internvideo"],"state":"official (archive's flag): 3 ran","n_ran":3,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/internvideo-general-video-foundation-models#ran","syntology_url":"https://syntology.ai/paper/2212.03191","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2212.03191"}}}},{"paper":"/paper/x-2-vlm-all-in-one-pre-trained-model-for","slug":"x-2-vlm-all-in-one-pre-trained-model-for","title":"X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks","date":"2022-11-22","arxiv_id":"2211.12402","rows_on_this_dataset":2,"code_links":2,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":6,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":1,"samples_ran_instrument_failed":2,"samples_unverified":3,"pointer_only_for_licence":6,"official":{"repos":["zengyan-97/x2-vlm"],"state":"official (archive's flag): 2 ran","n_ran":2,"n_constructed":0,"n_ran_no_instrument_failure":1,"n_unverified":3,"ran_from_kinds":["listed","official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/x-2-vlm-all-in-one-pre-trained-model-for#ran","syntology_url":"https://syntology.ai/paper/2211.12402","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2211.12402"}}}},{"paper":"/paper/expectation-maximization-contrastive-learning","slug":"expectation-maximization-contrastive-learning","title":"Expectation-Maximization Contrastive Learning for Compact Video-and-Language Representations","date":"2022-11-21","arxiv_id":"2211.11427","rows_on_this_dataset":3,"code_links":4,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":4,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":0,"samples_ran_instrument_failed":3,"samples_unverified":1,"pointer_only_for_licence":0,"official":{"repos":["jpthu17/emcl"],"state":"official (archive's flag): 1 ran","n_ran":1,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":1,"ran_from_kinds":["listed","official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/expectation-maximization-contrastive-learning#ran","syntology_url":"https://syntology.ai/paper/2211.11427","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2211.11427"}}}},{"paper":"/paper/clip-vip-adapting-pre-trained-image-text","slug":"clip-vip-adapting-pre-trained-image-text","title":"CLIP-ViP: Adapting Pre-trained Image-Text Model to Video-Language Representation Alignment","date":"2022-09-14","arxiv_id":"2209.06430","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":4,"samples_ran":4,"samples_constructed":1,"samples_ran_checked":1,"samples_ran_instrument_failed":3,"samples_unverified":0,"pointer_only_for_licence":4,"official":{"repos":["microsoft/xpretrain"],"state":"official (archive's flag): 4 ran","n_ran":4,"n_constructed":1,"n_ran_no_instrument_failure":1,"n_unverified":0,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/clip-vip-adapting-pre-trained-image-text#ran","syntology_url":"https://syntology.ai/paper/2209.06430","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2209.06430"}}}},{"paper":"/paper/ts2-net-token-shift-and-selection-transformer","slug":"ts2-net-token-shift-and-selection-transformer","title":"TS2-Net: Token Shift and Selection Transformer for Text-Video Retrieval","date":"2022-07-16","arxiv_id":"2207.07852","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":4,"samples_ran":4,"samples_constructed":4,"samples_ran_checked":4,"samples_ran_instrument_failed":0,"samples_unverified":0,"pointer_only_for_licence":4,"official":{"repos":["yuqi657/ts2_net"],"state":"official (archive's flag): 4 ran","n_ran":4,"n_constructed":4,"n_ran_no_instrument_failure":4,"n_unverified":0,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/ts2-net-token-shift-and-selection-transformer#ran","syntology_url":"https://syntology.ai/paper/2207.07852","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2207.07852"}}}},{"paper":"/paper/x-clip-end-to-end-multi-grained-contrastive","slug":"x-clip-end-to-end-multi-grained-contrastive","title":"X-CLIP: End-to-End Multi-grained Contrastive Learning for Video-Text Retrieval","date":"2022-07-15","arxiv_id":"2207.07285","rows_on_this_dataset":1,"code_links":3,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":2,"samples_ran":1,"samples_constructed":0,"samples_ran_checked":0,"samples_ran_instrument_failed":1,"samples_unverified":1,"pointer_only_for_licence":1,"official":{"repos":["xuguohai/X-CLIP"],"state":"official: harvested, nothing ran","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":1,"ran_from_kinds":["unlocated"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/x-clip-end-to-end-multi-grained-contrastive#ran","syntology_url":"https://syntology.ai/paper/2207.07285","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2207.07285"}}}},{"paper":"/paper/revealing-single-frame-bias-for-video-and","slug":"revealing-single-frame-bias-for-video-and","title":"Revealing Single Frame Bias for Video-and-Language Learning","date":"2022-06-07","arxiv_id":"2206.03428","rows_on_this_dataset":3,"code_links":2,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":12,"samples_ran":11,"samples_constructed":0,"samples_ran_checked":10,"samples_ran_instrument_failed":1,"samples_unverified":1,"pointer_only_for_licence":2,"official":{"repos":["jayleicn/singularity"],"state":"community repositories only","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["listed"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/revealing-single-frame-bias-for-video-and#ran","syntology_url":"https://syntology.ai/paper/2206.03428","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2206.03428"}}}},{"paper":"/paper/git-a-generative-image-to-text-transformer","slug":"git-a-generative-image-to-text-transformer","title":"GIT: A Generative Image-to-text Transformer for Vision and Language","date":"2022-05-27","arxiv_id":"2205.14100","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":21,"samples_ran":14,"samples_constructed":0,"samples_ran_checked":14,"samples_ran_instrument_failed":0,"samples_unverified":7,"pointer_only_for_licence":0,"official":{"repos":["microsoft/GenerativeImage2Text"],"state":"official (archive's flag): 14 ran","n_ran":14,"n_constructed":0,"n_ran_no_instrument_failure":14,"n_unverified":7,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/git-a-generative-image-to-text-transformer#ran","syntology_url":"https://syntology.ai/paper/2205.14100","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2205.14100"}}}},{"paper":"/paper/coca-contrastive-captioners-are-image-text","slug":"coca-contrastive-captioners-are-image-text","title":"CoCa: Contrastive Captioners are Image-Text Foundation Models","date":"2022-05-04","arxiv_id":"2205.01917","rows_on_this_dataset":1,"code_links":6,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":17,"samples_ran":10,"samples_constructed":5,"samples_ran_checked":10,"samples_ran_instrument_failed":0,"samples_unverified":7,"pointer_only_for_licence":2,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/coca-contrastive-captioners-are-image-text#ran","syntology_url":"https://syntology.ai/paper/2205.01917","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2205.01917"}}}},{"paper":"/paper/x-pool-cross-modal-language-video-attention","slug":"x-pool-cross-modal-language-video-attention","title":"X-Pool: Cross-Modal Language-Video Attention for Text-Video Retrieval","date":"2022-03-28","arxiv_id":"2203.15086","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":2,"samples_ran":1,"samples_constructed":1,"samples_ran_checked":1,"samples_ran_instrument_failed":0,"samples_unverified":1,"pointer_only_for_licence":2,"official":{"repos":["layer6ai-labs/xpool"],"state":"official (archive's flag): 1 ran","n_ran":1,"n_constructed":1,"n_ran_no_instrument_failure":1,"n_unverified":1,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/x-pool-cross-modal-language-video-attention#ran","syntology_url":"https://syntology.ai/paper/2203.15086","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2203.15086"}}}},{"paper":"/paper/disentangled-representation-learning-for-text","slug":"disentangled-representation-learning-for-text","title":"Disentangled Representation Learning for Text-Video Retrieval","date":"2022-03-14","arxiv_id":"2203.07111","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":12,"samples_ran":10,"samples_constructed":0,"samples_ran_checked":6,"samples_ran_instrument_failed":4,"samples_unverified":2,"pointer_only_for_licence":4,"official":{"repos":["foolwood/DRL"],"state":"official (archive's flag): 10 ran","n_ran":10,"n_constructed":0,"n_ran_no_instrument_failure":6,"n_unverified":2,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/disentangled-representation-learning-for-text#ran","syntology_url":"https://syntology.ai/paper/2203.07111","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2203.07111"}}}},{"paper":"/paper/bridgeformer-bridging-video-text-retrieval","slug":"bridgeformer-bridging-video-text-retrieval","title":"Bridging Video-text Retrieval with Multiple Choice Questions","date":"2022-01-13","arxiv_id":"2201.04850","rows_on_this_dataset":3,"code_links":2,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":24,"samples_ran":13,"samples_constructed":8,"samples_ran_checked":9,"samples_ran_instrument_failed":4,"samples_unverified":11,"pointer_only_for_licence":6,"official":{"repos":["tencentarc/mcq"],"state":"official (archive's flag): 2 ran","n_ran":2,"n_constructed":0,"n_ran_no_instrument_failure":1,"n_unverified":4,"ran_from_kinds":["listed","official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/bridgeformer-bridging-video-text-retrieval#ran","syntology_url":"https://syntology.ai/paper/2201.04850","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2201.04850"}}}},{"paper":"/paper/cross-modal-retrieval-with-querybank","slug":"cross-modal-retrieval-with-querybank","title":"Cross Modal Retrieval with Querybank Normalisation","date":"2021-12-23","arxiv_id":"2112.12777","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":3,"samples_ran":2,"samples_constructed":0,"samples_ran_checked":1,"samples_ran_instrument_failed":1,"samples_unverified":1,"pointer_only_for_licence":0,"official":{"repos":["ioanacroi/qb-norm"],"state":"official (archive's flag): 2 ran","n_ran":2,"n_constructed":0,"n_ran_no_instrument_failure":1,"n_unverified":1,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/cross-modal-retrieval-with-querybank#ran","syntology_url":"https://syntology.ai/paper/2112.12777","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2112.12777"}}}},{"paper":"/paper/nuwa-visual-synthesis-pre-training-for-neural","slug":"nuwa-visual-synthesis-pre-training-for-neural","title":"NÜWA: Visual Synthesis Pre-training for Neural visUal World creAtion","date":"2021-11-24","arxiv_id":"2111.12417","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":3,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":3,"samples_ran_instrument_failed":0,"samples_unverified":0,"pointer_only_for_licence":2,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/nuwa-visual-synthesis-pre-training-for-neural#ran","syntology_url":"https://syntology.ai/paper/2111.12417","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2111.12417"}}}},{"paper":"/paper/vatt-transformers-for-multimodal-self","slug":"vatt-transformers-for-multimodal-self","title":"VATT: Transformers for Multimodal Self-Supervised Learning from Raw Video, Audio and Text","date":"2021-04-22","arxiv_id":"2104.11178","rows_on_this_dataset":1,"code_links":5,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":8,"samples_ran":5,"samples_constructed":4,"samples_ran_checked":5,"samples_ran_instrument_failed":0,"samples_unverified":3,"pointer_only_for_licence":8,"official":{"repos":["google-research/google-research"],"state":"community repositories only","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["listed"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/vatt-transformers-for-multimodal-self#ran","syntology_url":"https://syntology.ai/paper/2104.11178","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2104.11178"}}}},{"paper":"/paper/clip4clip-an-empirical-study-of-clip-for-end","slug":"clip4clip-an-empirical-study-of-clip-for-end","title":"CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval","date":"2021-04-18","arxiv_id":"2104.08860","rows_on_this_dataset":4,"code_links":5,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":4,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":1,"samples_ran_instrument_failed":2,"samples_unverified":1,"pointer_only_for_licence":3,"official":{"repos":["ArrowLuo/CLIP4Clip"],"state":"official: harvested, nothing ran","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":1,"ran_from_kinds":["unlocated"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/clip4clip-an-empirical-study-of-clip-for-end#ran","syntology_url":"https://syntology.ai/paper/2104.08860","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2104.08860"}}}},{"paper":"/paper/frozen-in-time-a-joint-video-and-image","slug":"frozen-in-time-a-joint-video-and-image","title":"Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval","date":"2021-04-01","arxiv_id":"2104.00650","rows_on_this_dataset":3,"code_links":5,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":11,"samples_ran":10,"samples_constructed":0,"samples_ran_checked":7,"samples_ran_instrument_failed":3,"samples_unverified":1,"pointer_only_for_licence":3,"official":{"repos":["m-bain/frozen-in-time"],"state":"official (archive's flag): 10 ran","n_ran":10,"n_constructed":0,"n_ran_no_instrument_failure":7,"n_unverified":1,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/frozen-in-time-a-joint-video-and-image#ran","syntology_url":"https://syntology.ai/paper/2104.00650","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2104.00650"}}}},{"paper":"/paper/mdmmt-multidomain-multimodal-transformer-for","slug":"mdmmt-multidomain-multimodal-transformer-for","title":"MDMMT: Multidomain Multimodal Transformer for Video Retrieval","date":"2021-03-19","arxiv_id":"2103.10699","rows_on_this_dataset":2,"code_links":3,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":5,"samples_ran":4,"samples_constructed":0,"samples_ran_checked":0,"samples_ran_instrument_failed":4,"samples_unverified":1,"pointer_only_for_licence":5,"official":{"repos":["papermsucode/mdmmt"],"state":"official (archive's flag): 4 ran","n_ran":4,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":1,"ran_from_kinds":["official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/mdmmt-multidomain-multimodal-transformer-for#ran","syntology_url":"https://syntology.ai/paper/2103.10699","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2103.10699"}}}},{"paper":"/paper/multi-modal-transformer-for-video-retrieval","slug":"multi-modal-transformer-for-video-retrieval","title":"Multi-modal Transformer for Video Retrieval","date":"2020-07-21","arxiv_id":"2007.10639","rows_on_this_dataset":3,"code_links":1,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":7,"samples_ran":7,"samples_constructed":0,"samples_ran_checked":6,"samples_ran_instrument_failed":1,"samples_unverified":0,"pointer_only_for_licence":0,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/multi-modal-transformer-for-video-retrieval#ran","syntology_url":"https://syntology.ai/paper/2007.10639","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2007.10639"}}}},{"paper":"/paper/univilm-a-unified-video-and-language-pre","slug":"univilm-a-unified-video-and-language-pre","title":"UniVL: A Unified Video and Language Pre-Training Model for Multimodal Understanding and Generation","date":"2020-02-15","arxiv_id":"2002.06353","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":4,"samples_ran":1,"samples_constructed":0,"samples_ran_checked":0,"samples_ran_instrument_failed":1,"samples_unverified":3,"pointer_only_for_licence":1,"official":{"repos":["microsoft/UniVL"],"state":"official: harvested, nothing ran","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":1,"ran_from_kinds":["unlocated"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/univilm-a-unified-video-and-language-pre#ran","syntology_url":"https://syntology.ai/paper/2002.06353","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2002.06353"}}}},{"paper":"/paper/end-to-end-learning-of-visual-representations","slug":"end-to-end-learning-of-visual-representations","title":"End-to-End Learning of Visual Representations from Uncurated Instructional Videos","date":"2019-12-13","arxiv_id":"1912.06430","rows_on_this_dataset":1,"code_links":4,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":5,"samples_ran":4,"samples_constructed":0,"samples_ran_checked":2,"samples_ran_instrument_failed":2,"samples_unverified":1,"pointer_only_for_licence":2,"official":{"repos":["antoine77340/MIL-NCE_HowTo100M"],"state":"official (archive's flag): 1 ran","n_ran":1,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["listed","official"]},"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/end-to-end-learning-of-visual-representations#ran","syntology_url":"https://syntology.ai/paper/1912.06430","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1912.06430"}}}},{"paper":"/paper/use-what-you-have-video-retrieval-using","slug":"use-what-you-have-video-retrieval-using","title":"Use What You Have: Video Retrieval Using Representations From Collaborative Experts","date":"2019-07-31","arxiv_id":"1907.13487","rows_on_this_dataset":2,"code_links":3,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":5,"samples_ran":3,"samples_constructed":0,"samples_ran_checked":3,"samples_ran_instrument_failed":0,"samples_unverified":2,"pointer_only_for_licence":1,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/use-what-you-have-video-retrieval-using#ran","syntology_url":"https://syntology.ai/paper/1907.13487","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1907.13487"}}}},{"paper":"/paper/howto100m-learning-a-text-video-embedding-by","slug":"howto100m-learning-a-text-video-embedding-by","title":"HowTo100M: Learning a Text-Video Embedding by Watching Hundred Million Narrated Video Clips","date":"2019-06-07","arxiv_id":"1906.03327","rows_on_this_dataset":3,"code_links":4,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":1,"samples_ran":1,"samples_constructed":0,"samples_ran_checked":0,"samples_ran_instrument_failed":1,"samples_unverified":0,"pointer_only_for_licence":1,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/howto100m-learning-a-text-video-embedding-by#ran","syntology_url":"https://syntology.ai/paper/1906.03327","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1906.03327"}}}},{"paper":"/paper/a-joint-sequence-fusion-model-for-video","slug":"a-joint-sequence-fusion-model-for-video","title":"A Joint Sequence Fusion Model for Video Question Answering and Retrieval","date":"2018-08-07","arxiv_id":"1808.02559","rows_on_this_dataset":2,"code_links":2,"syntology":{"read_at":"2026-09-28T10:30:06+00:00","samples_harvested":2,"samples_ran":1,"samples_constructed":0,"samples_ran_checked":1,"samples_ran_instrument_failed":0,"samples_unverified":1,"pointer_only_for_licence":0,"official":null,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim.","sample_list":"/paper/a-joint-sequence-fusion-model-for-video#ran","syntology_url":"https://syntology.ai/paper/1808.02559","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1808.02559"}}}}],"record_sha256":"d1095737acf69e4d642f4dc61e473e6c815762e25a7046855bb667aa8795517a","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}