{"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/svbench","name":"SVBench","full_name":"Streaming Video Understanding Benchmark","description_markdown":"# **Dataset Card for SVBench**\r\n\r\nThis dataset card aims to provide a comprehensive overview of the SVBench dataset, including its purpose, structure, and sources. For details, see our [Project](https://yzy-bupt.github.io/SVBench/), [Paper](https://arxiv.org/abs/2502.10810) and [GitHub repository](https://github.com/yzy-bupt/SVBench).\r\n\r\n## **Dataset Details**\r\n\r\n### **Dataset Description**\r\n\r\nSVBench is the first benchmark specifically designed to evaluate **long-context streaming video understanding** through **temporal multi-turn question-answering (QA) chains**. It addresses the limitations of existing video understanding benchmarks by emphasizing **continuous temporal reasoning** over streaming video data.\r\n\r\nThe dataset includes:\r\n\r\n- **1,353 streaming videos** (average duration: 1–8 minutes)\r\n- **49,979 QA pairs** organized into **multi-turn dialogues**\r\n- **Temporal linkages** between QA chains to simulate real-world streaming scenarios\r\n\r\n**Languages**\r\n\r\n- **NLP**: English & Chinese (bilingual annotations)\r\n\r\n**License**\r\n\r\n- Apache-2.0\r\n\r\n**Dataset Sources**\r\n\r\n- [YT-Temporal-1B](https://rowanzellers.com/merlotreserve/#data)|[YouCook2](http://youcook2.eecs.umich.edu/download)|[MovieChat](https://huggingface.co/datasets/Enxin/MovieChat-1K-test/tree/main)|[Panda-70M](https://snap-research.github.io/Panda-70M/)|[Ego4D](https://ego4d-data.org/#download)|[ActivityNet](http://activity-net.org/download.html).\r\n\r\n## **Uses**\r\n\r\nDownload SVBench dataset from Hugging Face:\r\n\r\n\r\n\r\n```\r\ngit clone https://huggingface.co/yzy666/SVBench\r\n```\r\n\r\n\r\n\r\n**Intended Use**\r\n\r\n- Evaluate **long-context reasoning** capabilities of Large Vision-Language Models (LVLMs) in streaming video scenarios.\r\n- Benchmark **temporal understanding** through multi-turn dialogues.\r\n- Support research in **streaming video QA**, **activity forecasting**, and **interactive AI assistants**.\r\n\r\n**Direct Use**\r\n\r\n- Training or fine-tuning LVLMs for streaming video tasks.\r\n- Testing model robustness in handling **dynamic, time-sensitive queries**.\r\n- Comparative analysis of open-source vs. closed-source LVLMs.\r\n\r\n**Restrictions**\r\n\r\n- Videos and annotations are for **research purposes only**.\r\n- Users must comply with the original licenses of datasets.\r\n\r\n## **Dataset Structure**\r\n\r\n- **Folder Structure Tree:**\r\n\r\n```\r\nSVBench/\r\n├── Con/\r\n│   ├── Con_EN/\r\n│   └── Con_ZH/\r\n├── Dialogue/\r\n│   ├── Dialogue_EN/\r\n│   └── Dialogue_ZH/\r\n├── Meta/\r\n│   ├── Meta_EN/\r\n│   └── Meta_ZH/\r\n├── Path/\r\n├── Src/\r\n├── Streaming/\r\n│   ├── Streaming_EN/\r\n│   └── Streaming_ZH/\r\n├── Video/\r\n└── Your_Model_Name/\r\n    ├── dialogue/\r\n    │   └── --NDulaHyrE.json\r\n    └── streaming/\r\n        └── -4h8cuweoKo.json\r\n```\r\n\r\n- **Dataset Division:** \r\n\r\n- - Training Set: 42,605 QA pairs and 1,153 videos\r\n  - Testing Set: 7,374 QA pairs and 200 videos\r\n\r\n## **Data Fields**\r\n\r\n| **Key**                   | **Description**                                              |\r\n| ------------------------- | ------------------------------------------------------------ |\r\n| Video_Name                | Unique identifier or title of the video file.                |\r\n| Sort_of_Set               | Category or type of the dataset subset (e.g., \"Train\", \"Test\"). |\r\n| Path_of_QandA             | File path to the question-answer pairs.                      |\r\n| Path_of_Con               | Path to relationship files.                                  |\r\n| Path_of_StreamingPathData | Path to Q&A sequence for streaming evaluation. Each streaming path contains all Q&A sequences in streaming order within the path. |\r\n| Path_of_Dialogue          | Path to Q&A sequence for dialogue evaluation. Each dialogue contains all Q&A sequences in order within the dialogue. |\r\n| Path_of_Streaming         | Path to Q&A sequence for streaming evaluation represented only by serial numbers (e.g., the path [[0,0],[1,2],[2,3]...] indicates starting from the 1st question of the 1st chain, then proceeding to the 3rd question of the 2nd chain, and then to the 4th question of the 3rd chain, and so on). |\r\n| Path_of_Video             | Absolute file path to the raw video file.                    |\r\n| Video_Duration            | Total duration of the video in seconds.                      |\r\n| Source_of_Dataset         | Origin of the dataset.                                       |\r\n\r\n## **Leaderboard Submission**\r\n\r\nSubmit results via https://forms.gle/Rmi6u4WGhyEZ2X7g8.\r\n\r\n**Submission Instructions:**\r\n\r\n1. Save result files following the case structure under the SVBench/[Your_Model_Name] directory\r\n2. Compress the entire [Your_Model_Name] folder into a ZIP archive\r\n3. Upload the generated ZIP file through this submission portal\r\n\r\nImportant Notes:\r\n\r\n- Maintain the exact directory hierarchy: [Your_Model_Name]/[dialogue/streaming]/[Video_Name.json]\r\n- Ensure your model name follows exact capitalization (case-sensitive)\r\n- Package only the required result files (avoid including source code/executables)\r\n\r\n**Submission Form**\r\n\r\n![img](https://s2.loli.net/2025/03/16/AbhS7nB9Raf1uvk.png)\r\n\r\n## **Dataset Annotation**\r\n\r\nSee [GitHub repository](https://github.com/yzy-bupt/SVBench) for details.\r\n\r\nSemi-automated annotation using a hybrid approach:\r\n\r\n- Automatic QA Generation: Leveraging GPT-4 to generate dialogue chains based on video transcripts.\r\n- Human Verification: Annotators validate temporal links and refine QA pairs for consistency.\r\n- Temporal Linking: Artificial alignment of QA chains across video segments.\r\n- Quality Control: Inter-annotator agreement >90% for temporal links.\r\n\r\n## **Citation**\r\n\r\nIf you find our data useful, please consider citing our work!\r\n\r\n```\r\n@article{yang2025svbench,\r\n  title={SVBench: A Benchmark with Temporal Multi-Turn Dialogues for Streaming Video Understanding},\r\n  author={Yang, Zhenyu and Hu, Yuhang and Du, Zemin and Xue, Dizhan and Qian, Shengsheng and Wu, Jiahong and Yang, Fan and Dong, Weiming and Xu, Changsheng},\r\n  journal={arXiv preprint arXiv:2502.10810},\r\n  year={2025}\r\n}\r\n```","description_withheld":null,"homepage":"https://huggingface.co/datasets/yzy666/SVBench","introduced_date":"2025-02-15","introduced_date_note":null,"introduced_by":{"paper":"/paper/svbench-a-benchmark-with-temporal-multi-turn","title":"SVBench: A Benchmark with Temporal Multi-Turn Dialogues for Streaming Video Understanding","first_author":"Zhenyu Yang","url":null},"license":null,"modalities":[],"tasks":[{"name":"Video Understanding","url":"/task/video-understanding","datasets_with_task":"/datasets/task/video-understanding"}],"languages":[{"name":"English","url":"/datasets/language/english"},{"name":"Chinese","url":"/datasets/language/chinese"}],"variants":["SVBench"],"data_loaders":[],"num_papers_in_archive":2,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[],"papers":{"order":"date (newest first), then slug","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,"shown":0,"of":0,"counts":{"papers_with_a_benchmark_row":0,"with_a_code_link":0,"where_syntology_ran_a_sample":0,"not_listed_spam_title":0,"listed":0,"listed_where_code_ran":0,"where_syntology_ran_a_sample_split":{"with_a_run_with_no_instrument_failure":0,"every_run_a_failure_of_syntologys_instrument":0,"listed_with_a_run_with_no_instrument_failure":0,"listed_every_run_a_failure_of_syntologys_instrument":0,"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"},"items":[],"full_list":"papers_with_a_benchmark_row: every listed paper, in this order"},"papers_with_a_benchmark_row":[],"papers_where_code_ran":0,"papers_where_code_ran_first_page":null,"papers_where_code_ran_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.","pages":[],"row_route":"each paper's syntology object carries sample_list (this site's paper page, #ran), syntology_url (syntology.ai's page for the paper, only when that route probed live) and mcp (the agent call that lists the paper's harvested samples; not a verdict on the paper)","developers":"https://syntology.ai/developers","syntology_totals":{"read_at":"2026-10-05T00:36:11+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_constructed":0,"samples_ran_checked":0,"samples_ran_instrument_failed":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"papers_with_a_sample_that_ran":0,"papers_with_a_run_with_no_instrument_failure":0,"papers_where_every_run_was_an_instrument_failure":0,"note":"the per-paper counts of every listed paper (all pages), summed; not a rate"},"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)"},"since_archive_judged":{"papers":0,"papers_placed":0,"judged_by":{"model":"a model","models_as_recorded":[],"from":"lb_rows_run_verified.jsonl: each record's `model` field (the extractor's call that wrote decisions[])"},"note":"a model's judgement from the paper's own tables that they report on this leaderboard, not a result: no value from a paper that was not placed is printed or carried here; papers counts each paper once over these leaderboards","backfill":"8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read)","backfill_complete":false,"leaderboards":[]},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks.","record_sha256":"42c0c517e68c31b06c7e9b4149c29348d12f00c4c941f4f3577af927daa9509c","record_changed_at":"2026-10-05","record_changed_at_basis":"changed"}