{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T07:12:19Z","timestamp":1778051539563,"version":"3.51.4"},"reference-count":59,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,3,6]],"date-time":"2026-03-06T00:00:00Z","timestamp":1772755200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,3,6]],"date-time":"2026-03-06T00:00:00Z","timestamp":1772755200000},"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":[[2026,3,6]]},"DOI":"10.1109\/wacv61042.2026.00413","type":"proceedings-article","created":{"date-parts":[[2026,5,5]],"date-time":"2026-05-05T19:59:32Z","timestamp":1778011172000},"page":"4242-4252","source":"Crossref","is-referenced-by-count":0,"title":["Learning Compact Video Representations for Efficient Long-form Video Understanding in Large Multimodal Models"],"prefix":"10.1109","author":[{"given":"Yuxiao","family":"Chen","sequence":"first","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jue","family":"Wang","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhikang","family":"Zhang","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jingru","family":"Yi","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xu","family":"Zhang","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yang","family":"Zou","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhaowei","family":"Cai","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jianbo","family":"Yuan","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xinyu","family":"Li","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hao","family":"Yang","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Davide","family":"Modolo","sequence":"additional","affiliation":[{"name":"Amazon AGI"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW59228.2023.00484"},{"key":"ref2","article-title":"Token Merging: Your ViT but faster","volume-title":"ICLR","author":"Bolya"},{"key":"ref3","article-title":"A short note about kinetics-600","author":"Carreira","year":"2018"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-025-02597-y"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0614"},{"key":"ref6","article-title":"Videollama 2: Advancing spatial-temporal modeling and audio understanding in video-llms","author":"Cheng","year":"2024"},{"key":"ref7","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020"},{"key":"ref8","first-page":"5842","article-title":"The \u201c something something\u201d video database for learning and evaluating visual common sense","volume-title":"Proceedings of the IEEE International Conference on Computer Vision","author":"Goyal"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01842"},{"key":"ref10","article-title":"Mamba: Linear-time sequence modeling with selective state spaces","author":"Gu","year":"2023"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19833-5_6"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01723"},{"key":"ref13","first-page":"4651","article-title":"Perceiver: General perception with iterative attention","volume-title":"International conference on machine learning","author":"Jaegle"},{"key":"ref14","article-title":"Auto-encoding variational bayes","author":"Kingma","year":"2013"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.105"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.105"},{"key":"ref17","article-title":"The ava-kinetics localized human actions video dataset","author":"Li","year":"2020"},{"key":"ref18","article-title":"Llava-onevision: Easy visual task transfer","author":"Li","year":"2024"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4321-9"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02095"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00715"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72952-2_19"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72952-2_19"},{"key":"ref24","article-title":"Not all patches are what you need: Expediting vision transformers via token reorganizations","author":"Liang","year":"2022"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-emnlp.384"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.342"},{"key":"ref27","article-title":"Visual instruction tuning","volume":"36","author":"Liu","year":"2024","journal-title":"Advances in neural information processing systems"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/tnnls.2025.3610435"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.679"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2004"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01199"},{"key":"ref32","article-title":"Dinov2: Learning robust visual features without supervision","author":"Oquab","year":"2023"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1852"},{"key":"ref34","article-title":"DynamicViT: Efficient vision transformers with dynamic token sparsification","volume-title":"NeurIPS","author":"Rao"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.66"},{"issue":"599-607","key":"ref36","first-page":"6","article-title":"Learning internal representations by error propagation, parallel distributed processing, explorations in the microstructure of cognition","volume-title":"Biometrika","volume":"71","author":"Rumelhart","year":"1986"},{"key":"ref37","article-title":"Longvu: Spatiotemporal adaptive compression for long video-language understanding","author":"Shen","year":"2024"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01725"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2023.127063"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00130"},{"key":"ref41","article-title":"Neural discrete representation learning","volume":"30","author":"Den Oord","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19833-5_5"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00618"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00468"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72989-8_4"},{"key":"ref46","first-page":"4581","article-title":"Compressed vision for efficient video understanding","volume-title":"Proceedings of the Asian Conference on Computer Vision","author":"Wiles"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00192"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00965"},{"key":"ref49","article-title":"Slowfast-llava: A strong training-free baseline for video large language models","author":"Xu","year":"2024"},{"key":"ref50","article-title":"Qwen2 technical report","author":"Yang","year":"2024"},{"key":"ref51","article-title":"Vivim: a video vision mamba for medical video object segmentation","author":"Yang","year":"2024"},{"key":"ref52","article-title":"Clevrer: Collision events for video representation and reasoning","author":"Yi","year":"2019"},{"key":"ref53","article-title":"Language model beats diffusion\u2013tokenizer is key to visual generation","author":"Yu","year":"2023"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33019127"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.1209"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.1209"},{"key":"ref57","article-title":"Long context transfer from language to vision","author":"Zhang","year":"2024"},{"key":"ref58","author":"Zhang","year":"2024","journal-title":"Llava-next: A strong zero-shot video understanding model"},{"key":"ref59","article-title":"Mlvu: A comprehensive benchmark for multi-task long video understanding","author":"Zhou","year":"2024"}],"event":{"name":"2026 IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV)","location":"Tucson, AZ, USA","start":{"date-parts":[[2026,3,6]]},"end":{"date-parts":[[2026,3,10]]}},"container-title":["2026 IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11491838\/11491925\/11492707.pdf?arnumber=11492707","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T06:16:55Z","timestamp":1778048215000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11492707\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,6]]},"references-count":59,"URL":"https:\/\/doi.org\/10.1109\/wacv61042.2026.00413","relation":{},"subject":[],"published":{"date-parts":[[2026,3,6]]}}}