{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T09:59:42Z","timestamp":1777888782187,"version":"3.51.4"},"reference-count":123,"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\/iccv51701.2025.02177","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"23453-23465","source":"Crossref","is-referenced-by-count":0,"title":["Bringing RNNs Back to Efficient Open-Ended Video Understanding"],"prefix":"10.1109","author":[{"given":"Weili","family":"Xu","sequence":"first","affiliation":[{"name":"Zhejiang University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Enxin","family":"Song","sequence":"additional","affiliation":[{"name":"Zhejiang University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wenhao","family":"Chai","sequence":"additional","affiliation":[{"name":"University of Washington"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xuexiang","family":"Wen","sequence":"additional","affiliation":[{"name":"Zhejiang University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tian","family":"Ye","sequence":"additional","affiliation":[{"name":"HKUST (GZ)"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gaoang","family":"Wang","sequence":"additional","affiliation":[{"name":"Zhejiang University,Shanghai AI Lab"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","author":"Achiam","year":"2023","journal-title":"Gpt-4 technical report"},{"key":"ref2","author":"Hasan","year":"2024","journal-title":"Hired: Attention-guided token dropping for efficient inference of high-resolution vision-language models in resource-constrained environments"},{"key":"ref3","author":"Ataallah","year":"2024","journal-title":"Infinibench: A comprehensive benchmark for large multimodal models in very long video understanding"},{"key":"ref4","author":"Bai","year":"2025","journal-title":"Qwen2. 5-vl technical report"},{"key":"ref5","author":"Beyer","year":"2024","journal-title":"Paligemma: A versatile 3b vlm for transfer"},{"key":"ref6","author":"Bharadwaj","year":"2024","journal-title":"Vane-bench: Video anomaly evaluation benchmark for conversational 1 mms"},{"key":"ref7","article-title":"Token merging: Your ViT but faster","volume-title":"International Conference on Learning Representations","author":"Bolya","year":"2023"},{"key":"ref8","first-page":"961","article-title":"Niebles","volume-title":"Proceedings of the ieee conference on computer vision and pattern recognition","author":"Caba Heilbron","year":"2015"},{"key":"ref9","author":"Cai","year":"2024","journal-title":"Lee. Matryoshka multimodal models"},{"key":"ref10","author":"Chai","year":"2025","journal-title":"View transformer layers from online optimization perspective"},{"key":"ref11","author":"Chai","year":"2024","journal-title":"Manning. Auroracap: Efficient, performant video detailed captioning and a new benchmark"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-025-02597-y"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0614"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73004-7_2"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3627546\/mm1"},{"key":"ref16","author":"Chen","year":"2024","journal-title":"Longvila: Scaling long-context visual language models for long videos"},{"key":"ref17","author":"Chen","year":"2024","journal-title":"Stuffed mamba: State collapse and state capacity of rnn-based long-context modeling"},{"key":"ref18","author":"Chen","year":"2023","journal-title":"InternVL: Scaling up vision foundation models and aligning for generic visual-linguistic tasks"},{"key":"ref19","author":"Chen","year":"2024","journal-title":"Expanding performance boundaries of open-source multimodal models with model, data, and testtime scaling"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4231-5"},{"key":"ref21","author":"Cheng","year":"2024","journal-title":"Videollama 2: Advancing spatialtemporal modeling and audio understanding in video-llms"},{"key":"ref22","author":"Dao","year":"2024","journal-title":"Transformers are ssms: Generalized models and efficient algorithms through structured state space duality"},{"key":"ref23","author":"De","year":"2024","journal-title":"Griffin: Mixing gated linear recurrences with local attention for efficient language models"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/MWSCAS.2017.8053243"},{"key":"ref25","author":"Du","year":"2024","journal-title":"Towards event-oriented long video understanding"},{"key":"ref26","author":"Duan","year":"2024","journal-title":"Vision-rwkv: Efficient and scalable visual perception with rwkv-like architectures"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72670-5_5"},{"key":"ref28","author":"Feng","year":"2024","journal-title":"Were rnns all we needed?"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.02245"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-24797-2_4"},{"key":"ref31","author":"Gu","year":"2023","journal-title":"Mamba: Linear-time sequence modeling with selective state spaces"},{"key":"ref32","author":"Han","year":"2023","journal-title":"Shot2story20k: A new benchmark for comprehensive understanding of multi-shot videos"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01282"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01553"},{"key":"ref35","author":"Hou","year":"2024","journal-title":"Yu. Visualrwkv: Exploring recurrent neural networks for visual language models"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72640-8_5"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01300"},{"key":"ref38","author":"Kang","year":"2024","journal-title":"Exploring enhanced contextual information for video-level object tracking"},{"key":"ref39","first-page":"5156","article-title":"Transformers are rnns: Fast autoregressive transformers with linear attention","volume-title":"International conference on machine learning","author":"Katharopoulos","year":"2020"},{"key":"ref40","author":"Ke","year":"2020","journal-title":"Rethinking positional encoding in language pre-training"},{"key":"ref41","author":"Lee","year":"2024","journal-title":"Video token merging for long-form video understanding"},{"key":"ref42","author":"Li","year":"2025","journal-title":"Minimax-01: Scaling foundation models with lightning attention"},{"key":"ref43","author":"Li","year":"2024","journal-title":"Lmms-eval: Accelerating the development of large multimoal models"},{"key":"ref44","author":"Li","year":"2024","journal-title":"Llava-onevision: Easy visual task transfer"},{"key":"ref45","author":"Li","year":"2024","journal-title":"Aria: An open multimodal native mixture-of-experts model"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02095"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73347-5_14"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00572"},{"key":"ref49","first-page":"75","article-title":"Mamband: Selective state space modeling for multi-dimensional data","volume-title":"European Conference on Computer Vision","author":"Li","year":"2024"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72952-2_19"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72952-2_19"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.342"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02520"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref55","first-page":"26296","article-title":"Lee","volume-title":"Proceedings of the ieee\/CVF Conference on Computer Vision and Pattern Recognition","author":"Liu","year":"2024"},{"key":"ref56","author":"Liu","year":"2024","journal-title":"Llava-next: Improved reasoning, ocr, and world knowledge"},{"key":"ref57","article-title":"Visual instruction tuning","volume":"36","author":"Liu","year":"2024","journal-title":"Advances in neural information processing systems"},{"key":"ref58","author":"Liu","year":"2024","journal-title":"Chen. Et bench: Towards open-ended event-level video-language understanding"},{"key":"ref59","author":"Lu","year":"2024","journal-title":"Videomambapro: A leap forward for mamba in video understanding"},{"key":"ref60","author":"Ma","year":"2023","journal-title":"Vista-llama: Reliable video narrator via equal distance to visual tokens"},{"key":"ref61","author":"Ma","year":"2024","journal-title":"Video token sparsification for efficient multimodal 11 ms in autonomous driving"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.679"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2004"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72698-9_1"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.936"},{"key":"ref66","author":"Peng","year":"2024","journal-title":"Eagle and finch: Rwkv with matrix-valued states and dynamic recurrence"},{"key":"ref67","author":"Peng","year":"2025","journal-title":"Rwkv-7 \u201cgoose\u201d with expressive dynamic state evolution"},{"key":"ref68","author":"Qiao","year":"2024","journal-title":"Vl-mamba: Exploring state space models for multimodal learning"},{"key":"ref69","author":"Qin","year":"2024","journal-title":"Lightning attention-2: A free lunch for handling unlimited sequence lengths in large language models"},{"key":"ref70","author":"Qin","year":"2024","journal-title":"Various lengths, constant speed: Efficient language modeling with lightning attention"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1145\/3180308.3180329"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/78.650093"},{"key":"ref73","author":"Shang","year":"2024","journal-title":"Llava-prumerge: Adaptive token reduction for efficient large multimodal models"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11941"},{"key":"ref75","author":"Shen","year":"2024","journal-title":"Longvu: Spatiotemporal adaptive compression for long video-language understanding"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.02436"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.01725"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3604614"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/iccvw69036.2025.00642"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2023.127063"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01289"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.01769"},{"key":"ref83","author":"Touvron","year":"2023","journal-title":"Llama: Open and efficient foundation language models"},{"key":"ref84","article-title":"Attention is all you need","author":"Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref85","author":"Wang","year":"2024","journal-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution"},{"key":"ref86","author":"Wang","year":"2024","journal-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution"},{"key":"ref87","author":"Wang","year":"2023","journal-title":"Cogvlm: Visual expert for pretrained language models"},{"key":"ref88","author":"Wang","year":"2024","journal-title":"Lvbench: An extreme long video understanding benchmark"},{"key":"ref89","first-page":"4581","volume-title":"Proceedings of the ieee\/CVF International Conference on Computer Vision","author":"Wang","year":"2019"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72989-8_4"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73414-4_26"},{"key":"ref92","author":"Wu","year":"2024","journal-title":"Longvideobench: A benchmark for long-context interleaved video-language understanding"},{"key":"ref93","author":"Xu","year":"2024","journal-title":"Motionbank: A large-scale video motion benchmark with disentangled rule-based annotations"},{"key":"ref94","author":"Xu","year":"2024","journal-title":"Pllava: Parameter-free llava extension from images to videos for video dense captioning"},{"key":"ref95","author":"Xue","year":"2024","journal-title":"Longvila: Scaling long-context visual language models for long videos"},{"key":"ref96","author":"Xue","year":"2024","journal-title":"xgen-mm (blip-3): A family of open large multimodal models"},{"key":"ref97","author":"Yang","year":"2024","journal-title":"Qwen2 technical report"},{"key":"ref98","author":"Yang","year":"2023","journal-title":"Gated linear attention transformers with hardware-efficient training"},{"key":"ref99","author":"Yang","year":"2024","journal-title":"Gated delta networks: Improving mamba2 with delta rule"},{"key":"ref100","first-page":"115491","article-title":"Parallelizing linear transformers with the delta rule over sequence length","volume":"37","author":"Yang","year":"2025","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref101","author":"Ye","year":"2023","journal-title":"mplug-owl: Modularization empowers large language models with multimodality"},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33019127"},{"key":"ref103","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.836"},{"key":"ref104","author":"Zeng","year":"2024","journal-title":"Timesuite: Improving mllms for long video understanding via grounded tuning"},{"key":"ref105","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01100"},{"key":"ref106","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.1209"},{"key":"ref107","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01295"},{"key":"ref108","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-demo.49"},{"key":"ref109","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-naacl.51"},{"key":"ref110","author":"Zhang","year":"2024","journal-title":"Internlm-xcomposer-2.5: A versatile large vision language model supporting long-contextual input and output"},{"key":"ref111","author":"Zhang","year":"2024","journal-title":"Long context transfer from language to vision"},{"key":"ref112","author":"Zhang","year":"2024","journal-title":"Long context transfer from language to vision"},{"key":"ref113","author":"Zhang","year":"2024","journal-title":"[cls] attention is all you need for trainingfree visual token pruning: Make vlm inference faster"},{"key":"ref114","author":"Zhang","year":"2023","journal-title":"Llama-adapter: Efficient fine-tuning of language models with zero-init attention"},{"key":"ref115","article-title":"1nftybench: Extending long context evaluation beyond 100k tokens","author":"Zhang","year":"2024","journal-title":"ACL (1)"},{"key":"ref116","author":"Zhang","year":"2024","journal-title":"Llava-next: A strong zero-shot video understanding model"},{"key":"ref117","author":"Zhang","year":"2024","journal-title":"Gated slot attention for efficient lineartime sequence modeling"},{"key":"ref118","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01245"},{"key":"ref119","author":"Zhou","year":"2024","journal-title":"Mlvu: A comprehensive benchmark for multi-task long video understanding"},{"key":"ref120","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01727"},{"key":"ref121","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.940"},{"key":"ref122","article-title":"Vision mamba: Efficient visual representation learning with bidirectional state space model","volume-title":"International Conference on Machine Learning","author":"Zhu","year":"2024"},{"key":"ref123","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.01760"}],"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\/11445866.pdf?arnumber=11445866","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T04:57:23Z","timestamp":1777611443000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11445866\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":123,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.02177","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}