{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T10:22:49Z","timestamp":1777890169803,"version":"3.51.4"},"reference-count":80,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.01957","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"21059-21069","source":"Crossref","is-referenced-by-count":0,"title":["Flash-Vstream: Efficient Real-Time Understanding for Long Video Streams"],"prefix":"10.1109","author":[{"given":"Haoji","family":"Zhang","sequence":"first","affiliation":[{"name":"Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yiqin","family":"Wang","sequence":"additional","affiliation":[{"name":"Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yansong","family":"Tang","sequence":"additional","affiliation":[{"name":"Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yong","family":"Liu","sequence":"additional","affiliation":[{"name":"Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiashi","family":"Feng","sequence":"additional","affiliation":[{"name":"ByteDance Inc."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaojie","family":"Jin","sequence":"additional","affiliation":[{"name":"Beijing Jiaotong University"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/tip.2025.3639996"},{"key":"ref3","first-page":"2527","article-title":"Memory consolidation enables long-context video understanding","author":"Balazevic","year":"2024","journal-title":"In ICML"},{"key":"ref4","first-page":"33","article-title":"Language models are few-shot learners","author":"Brown","year":"2020","journal-title":"NeurIPS"},{"key":"ref5","article-title":"A memory-network based solution for multivariate timeseries forecasting","author":"Chang","year":"2018","journal-title":"arXiv preprint arXiv"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2019.2909473"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01742"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0614"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19815-1_37"},{"key":"ref10","article-title":"Videollama 2: Advancing spatialtemporal modeling and audio understanding in video-llms","author":"Cheng","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2142"},{"key":"ref12","article-title":"Flashattention-2: Faster attention with better parallelism and work partitioning","author":"Dao","year":"2024","journal-title":"ICLR"},{"key":"ref13","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2021","journal-title":"ICLR"},{"key":"ref14","article-title":"The llama 3 herd of models","author":"Dubey","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref15","first-page":"226","article-title":"A density-based algorithm for discovering clusters in large spatial databases with noise","author":"Ester","year":"1996","journal-title":"KDD"},{"key":"ref16","first-page":"24108","article-title":"Video-mme: The first-ever comprehensive evaluation benchmark of multi-modal 11 ms in video analysis","author":"Fu","year":"2025","journal-title":"In CVPR"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3581783.3612120"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01419"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-023-01891-x"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01535"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3117837"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00508"},{"key":"ref23","article-title":"LoRA: Low-rank adaptation of large language models","author":"Hu","year":"2022","journal-title":"ICLR"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58548-8_41"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01300"},{"key":"ref26","article-title":"Llava-onevision: Easy visual task transfer","author":"Li","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref27","first-page":"12888","article-title":"Blip: Bootstrapping language-image pre-training for unified visionlanguage understanding and generation","author":"Li","year":"2022","journal-title":"In ICML"},{"key":"ref28","first-page":"19730","article-title":"Blip2: Bootstrapping language-image pre-training with frozen image encoders and large language models","author":"Li","year":"2023","journal-title":"In ICML"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02095"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72952-2_19"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-025-02620-2"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19818-2_27"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00332"},{"key":"ref36","article-title":"Oryx mllm: On-demand spatialtemporal understanding at arbitrary resolution","author":"Liu","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref37","first-page":"26425","article-title":"Soc: Semanticassisted object cluster for referring video object segmentation","author":"Luo","year":"2023","journal-title":"NeurIPS"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.2976014"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01249"},{"key":"ref40","article-title":"Some methods for classification and analysis of multivariate observations","volume-title":"Proceedings of 5-th Berkeley Symposium on Mathematical Statistics and Probability","author":"MacQueen","year":"1967"},{"issue":"4621246244","key":"ref41","article-title":"Egoschema: A diagnostic benchmark for very longform video language understanding","volume":"36","author":"Mangalam","year":"2023","journal-title":"NeurIPS"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2019.2960536"},{"key":"ref43","first-page":"27730","article-title":"Training language models to follow instructions with human feedback","author":"Ouyang","year":"2022","journal-title":"NeurIPS"},{"key":"ref44","first-page":"119336","article-title":"Streaming long video understanding with large language models","volume":"37","author":"Qian","year":"2025","journal-title":"NeurIPS"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01357"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA57147.2024.10610216"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01725"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2023.127063"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.43"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i5.16564"},{"key":"ref51","article-title":"Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context","author":"Team","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref52","article-title":"Kimi-vl technical report","author":"Team","year":"2025","journal-title":"arXiv preprint arXiv"},{"key":"ref53","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref54","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref55","article-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution","author":"Wang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref56","article-title":"Lvbench: An extreme long video understanding benchmark","author":"Wang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref57","article-title":"Retake: Reducing temporal and knowledge redundancy for long video understanding","author":"Wang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.01594"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01943"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19772-7_14"},{"key":"ref61","article-title":"Hierarchical memory for long video qa","author":"Wang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.76"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2024.3514654"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i8.32880"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02694"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58621-8_7"},{"key":"ref67","first-page":"109922","article-title":"Videollm-mod: Efficient videolanguage streaming with mixture-of-depths vision computation","volume":"37","author":"Wu","year":"2024","journal-title":"NeurIPS"},{"key":"ref68","article-title":"Longvila: Scaling long-context visual language models for long videos","author":"Xue","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2019.2946823"},{"key":"ref70","article-title":"Qwen2 technical report","author":"Yang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2024.3468640"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02325"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02777"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.52202\/075280-3354"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.297"},{"key":"ref76","article-title":"Long context transfer from language to vision","author":"Zhang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref77","article-title":"Video instruction tuning with synthetic data","author":"Zhang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref78","article-title":"Mlvu: A comprehensive benchmark for multi-task long video understanding","author":"Zhou","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01727"},{"key":"ref80","article-title":"Instarevive: One-step image enhancement via dynamic score matching","author":"Zhu","year":"2025","journal-title":"arXiv preprint arXiv"}],"event":{"name":"2025 IEEE\/CVF International Conference on Computer Vision (ICCV)","location":"Honolulu, HI, USA","start":{"date-parts":[[2025,10,19]]},"end":{"date-parts":[[2025,10,25]]}},"container-title":["2025 IEEE\/CVF International Conference on Computer Vision (ICCV)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11443115\/11443287\/11444205.pdf?arnumber=11444205","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:29:29Z","timestamp":1777613369000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444205\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":80,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01957","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}