{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T20:21:39Z","timestamp":1792786899957,"version":"4.5.0"},"reference-count":125,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"11","license":[{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Intelligence Advanced Research Projects Activity","award":["140D0423C0074"],"award-info":[{"award-number":["140D0423C0074"]}],"id":[{"id":"https:\/\/ror.org\/04hm2xn53","id-type":"ROR","asserted-by":"crossref"}]},{"name":"Institute of Information & Communications Technology Planning & Evaluation","id":[{"id":"https:\/\/ror.org\/01g0hqq23","id-type":"ROR","asserted-by":"crossref"}]},{"name":"Korean Government","award":["RS-2024-00457882"],"award-info":[{"award-number":["RS-2024-00457882"]}]},{"name":"National AI Research Lab Project"},{"name":"National Natural Science Foundation of China","award":["62676293"],"award-info":[{"award-number":["62676293"]}],"id":[{"id":"https:\/\/ror.org\/01h0zpd94","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2026,11]]},"DOI":"10.1109\/tpami.2026.3720379","type":"journal-article","created":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T19:03:24Z","timestamp":1785870204000},"page":"14922-14936","source":"Crossref","is-referenced-by-count":24,"title":["Sa2VA: Marrying SAM2 With MLLM for Dense Grounded Understanding of Images and Videos"],"prefix":"10.1109","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9770-7720","authenticated-orcid":false,"given":"Haobo","family":"Yuan","sequence":"first","affiliation":[{"name":"University of California","place":["Merced, CA, USA"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0550-8247","authenticated-orcid":false,"given":"Xiangtai","family":"Li","sequence":"additional","affiliation":[{"name":"Bytedance Seed","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7390-2409","authenticated-orcid":false,"given":"Tao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Bytedance Seed","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-9455-1360","authenticated-orcid":false,"given":"Yueyi","family":"Sun","sequence":"additional","affiliation":[{"name":"Peking University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6744-0671","authenticated-orcid":false,"given":"Zilong","family":"Huang","sequence":"additional","affiliation":[{"name":"Bytedance Seed","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-7178-5358","authenticated-orcid":false,"given":"Shilin","family":"Xu","sequence":"additional","affiliation":[{"name":"Peking University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3088-1481","authenticated-orcid":false,"given":"Shunping","family":"Ji","sequence":"additional","affiliation":[{"name":"Wuhan University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8735-2516","authenticated-orcid":false,"given":"Yunhai","family":"Tong","sequence":"additional","affiliation":[{"name":"Peking University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lu","family":"Qi","sequence":"additional","affiliation":[{"name":"Wuhan University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6843-0064","authenticated-orcid":false,"given":"Jiashi","family":"Feng","sequence":"additional","affiliation":[{"name":"Bytedance Seed","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4848-2304","authenticated-orcid":false,"given":"Ming-Hsuan","family":"Yang","sequence":"additional","affiliation":[{"name":"University of California","place":["Merced, CA, USA"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01236"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2291"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72633-0_6"},{"key":"ref4","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023"},{"key":"ref5","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1177\/0032258x8005300111"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.279"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01438"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW69036.2025.00197"},{"key":"ref10","article-title":"Animate-a-story: Storytelling with retrieval-augmented video generation","author":"He","year":"2023"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3501"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-025-10041-x"},{"key":"ref13","first-page":"25105","article-title":"VideoPoet: A large language model for zero-shot video generation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Kondratyuk","year":"2024"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2024.3398538"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.02121"},{"key":"ref16","article-title":"Zero-shot video editing using off-the-shelf image diffusion models","author":"Wang","year":"2023"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00821"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2022.3187278"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10161251"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/WACV45572.2020.9093335"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2021.3108405"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00304"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02640"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00124"},{"key":"ref25","article-title":"Sam 2: Segment anything in images and videos","author":"Ravi","year":"2024"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-demo.49"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72952-2_19"},{"key":"ref28","article-title":"Free video-LLM: Prompt-guided visual perception for efficient training-free video LLMs","author":"Han","year":"2024"},{"key":"ref29","article-title":"Llava-next: Improved reasoning, ocr, and world knowledge","author":"Liu","year":"2024"},{"key":"ref30","article-title":"Aria: An open multimodal native mixture-of-experts model","author":"Li","year":"2024"},{"issue":"10","key":"ref31","volume":"68","author":"Li","journal-title":"Sci. China Inf. Sci."},{"key":"ref32","article-title":"Llava-onevision: Easy visual task transfer","volume":"2025","author":"Li","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.52202\/079017-1823"},{"key":"ref34","article-title":"Grounded chain-of-thought for multimodal large language models","author":"Wu","year":"2025"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i7.32718"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02664"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01773"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0219"},{"key":"ref39","first-page":"22062","article-title":"Memory-space visual prompting for efficient vision-language fine-tuning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Jie"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02301"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4231-5"},{"key":"ref43","article-title":"PLLaVA: Parameter-free LLaVA extension from images to videos for video dense captioning","author":"Xu","year":"2024"},{"key":"ref44","first-page":"12914","article-title":"Maki ng large multimodal models understand arbitrary visual prompts","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Cai","year":"2024"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00915"},{"key":"ref46","first-page":"8931","article-title":"HyperSeg: Towards universal visual segmentation with large language model","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Wei","year":"2025"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.01878"},{"key":"ref48","article-title":"Internvl3: Exploring advanced training and test-time recipes for open-source multimodal models","author":"Zhu","year":"2025"},{"key":"ref49","article-title":"Qwen3-omni technical report","author":"Xu","year":"2025"},{"key":"ref50","article-title":"Qwen2. 5-vl technical report","author":"Bai","year":"2025"},{"key":"ref51","first-page":"12888","article-title":"Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Li","year":"2022"},{"key":"ref52","first-page":"19730","article-title":"Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Li","year":"2023"},{"key":"ref53","article-title":"Pixel-bert: Aligning image pixels with text by deep multi-modal transformers","author":"Huang","year":"2020"},{"key":"ref54","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Brown","year":"2020"},{"key":"ref55","article-title":"Qwen-vl: A versatile vision-language model for understanding, localization, text reading, and beyond","author":"Bai","year":"2023"},{"key":"ref56","article-title":"Gemini: a family of highly capable multimodal models","author":"Team","year":"2023"},{"key":"ref57","article-title":"LLaVA-interactive: An all-in-one demo for image chat, segmentation, generation and editing","author":"Chen","year":"2023"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00686"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72658-3_13"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.20"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.52202\/085713-4899"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2771"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72775-7_2"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-024-02214-4"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02491"},{"key":"ref66","article-title":"Internlm-xcomposer: A vision-language large model for advanced text-image comprehension and composition","author":"Zhang","year":"2023"},{"key":"ref67","article-title":"Internlm-xcomposer2: Mastering free-form text-image composition and comprehension in vision-language large model","author":"Dong","year":"2024"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.52202\/079017-1348"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.342"},{"key":"ref70","article-title":"Generalizable entity grounding via assistance of large language model","author":"Qi","year":"2024"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/3DV66043.2025.00112"},{"key":"ref72","article-title":"Pixel-sail: Single transformer for pixel-grounded understanding","author":"Zhang","year":"2025"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01471"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2025.3552694"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00142"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02259"},{"key":"ref77","first-page":"123","article-title":"Video object segmentation with language referring expressions","volume-title":"Proc. Asian Conf. Comput. Vis.","author":"Khoreva","year":"2018"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00254"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01188"},{"key":"ref80","first-page":"74","article-title":"PSALM: Pixelwise segmentation with large multi-modal model","volume-title":"Proc. Eur. Conf. Comput. Vis.","author":"Zhang","year":"2024"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_13"},{"key":"ref82","article-title":"Deformable detr: Deformable transformers for end-to-end object detection","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zhu"},{"key":"ref83","first-page":"239","article-title":"Context-guided spatial feature reconstruction for efficient semantic segmentation","volume-title":"Proc. Eur. Conf. Comput. Vis.","author":"Ni","year":"2024"},{"key":"ref84","article-title":"Deco: Unleashing the potential of convnets for query-based detection and segmentation","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Chen","year":"2025"},{"key":"ref85","first-page":"446","article-title":"Improving video segmentation via dynamic anchor queries","volume-title":"Proc. Eur. Conf. Comput. Vis.","author":"Zhou","year":"2024"},{"key":"ref86","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00370"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2222"},{"key":"ref88","first-page":"513","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit. Workshops","author":"Chen"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01349"},{"key":"ref90","first-page":"13877","article-title":"Tube-link: A flexible cross tube baseline for universal video segmentation","volume-title":"Proc. IEEE\/CVF Int. Conf. Comput. Vis.","author":"Li","year":"2023"},{"key":"ref91","first-page":"13352","article-title":"Video instance segmentation using inter-frame communication transformers","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Hwang","year":"2021"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19818-2_30"},{"key":"ref93","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i3.32338"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2025.3562930"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01353"},{"key":"ref96","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00135"},{"key":"ref97","article-title":"Expanding performance boundaries of open-source multimodal models with model, data, and test-time scaling","author":"Chen","year":"2024"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1086"},{"key":"ref99","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46475-6_5"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58555-6_13"},{"key":"ref101","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01263"},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02245"},{"key":"ref103","first-page":"89098","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Fang"},{"key":"ref104","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref105","article-title":"mPLUG-Owl3: Towards long image-sequence understanding in multi-modal large language models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Ye","year":"2025"},{"key":"ref106","article-title":"LaSagnA: Language-based segmentation assistant for complex queries","author":"Wei","year":"2024"},{"key":"ref107","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72970-6_8"},{"key":"ref108","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.679"},{"key":"ref109","article-title":"Xtuner: A toolkit for efficiently fine-tuning LLM","author":"Contributors"},{"key":"ref110","doi-asserted-by":"publisher","DOI":"10.1145\/3664647.3685520"},{"key":"ref111","doi-asserted-by":"publisher","DOI":"10.1109\/iccv51070.2023.00240"},{"key":"ref112","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_15"},{"key":"ref113","first-page":"27056","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Chen"},{"key":"ref114","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00913"},{"key":"ref115","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0182"},{"key":"ref116","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01762"},{"key":"ref117","article-title":"Infinity-mm: Scaling multimodal performance with large-scale and high-quality instruction data","author":"Gu","year":"2024"},{"key":"ref118","article-title":"PG-video-LLaVA: Pixel grounding large video-language models","author":"Munasinghe","year":"2023"},{"key":"ref119","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00492"},{"key":"ref120","volume":"173","author":"Zhang","journal-title":"Image Vis. Comput."},{"key":"ref121","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00311"},{"key":"ref122","first-page":"207","article-title":"Grit: A generative region-to-text transformer for object understanding","volume-title":"Proc. Eur. Conf. Comput. Vis.","author":"Wu","year":"2024"},{"key":"ref123","article-title":"Kosmos-2: Grounding multimodal large language models to the world","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Peng","year":"2023"},{"key":"ref124","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW67362.2025.00251"},{"key":"ref125","article-title":"LSVOS 2025 challenge report: Recent advances in complex video object segmentation","author":"Liu","year":"2025"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11723300\/11640960.pdf?arnumber=11640960","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T19:14:56Z","timestamp":1791486896000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11640960\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,11]]},"references-count":125,"journal-issue":{"issue":"11"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2026.3720379","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"value":"0162-8828","type":"print"},{"value":"2160-9292","type":"electronic"},{"value":"1939-3539","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,11]]}}}