{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T10:00:35Z","timestamp":1777888835434,"version":"3.51.4"},"reference-count":45,"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\/iccv51701.2025.02219","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"23935-23944","source":"Crossref","is-referenced-by-count":0,"title":["$\\mathcal{D}$-Attn: Decomposed Attention for Large Vision-and-Language Models"],"prefix":"10.1109","author":[{"given":"Chia-Wen","family":"Kuo","sequence":"first","affiliation":[{"name":"ByteDance Intelligent Creation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sijie","family":"Zhu","sequence":"additional","affiliation":[{"name":"ByteDance Intelligent Creation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fan","family":"Chen","sequence":"additional","affiliation":[{"name":"ByteDance Intelligent Creation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaohui","family":"Shen","sequence":"additional","affiliation":[{"name":"ByteDance Intelligent Creation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Longyin","family":"Wen","sequence":"additional","affiliation":[{"name":"ByteDance Intelligent Creation"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Phi-3 technical report: A highly capable language model locally on your phone","author":"Abdin","year":"2024","journal-title":"arXiv preprint"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1723"},{"key":"ref3","article-title":"Openflamingo: An opensource framework for training large autoregressive visionlanguage models","author":"Awadalla","year":"2023","journal-title":"arXiv preprint"},{"key":"ref4","article-title":"Training a helpful and harmless assistant with reinforcement learning from human feedback","author":"Bai","year":"2022","journal-title":"arXiv preprint"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72643-9_22"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0850"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2142"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1189"},{"key":"ref9","article-title":"Molmo and pixmo: Open weights and open data for state-of-the-art multimodal models","author":"Deitke","year":"2024","journal-title":"arXiv preprint"},{"key":"ref10","article-title":"The llama 3 herd of models","author":"Dubey","year":"2024","journal-title":"arXiv preprint"},{"key":"ref11","volume-title":"Mme: A compre-hensive evaluation benchmark for multimodal large language models","author":"Fu","year":"2024"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.670"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00686"},{"key":"ref14","article-title":"Mistral 7b","author":"Jiang","year":"2023","journal-title":"arXiv preprint"},{"key":"ref15","article-title":"Obelics: An open web-scale filtered dataset of interleaved image-text documents","volume":"36","author":"Lauren\u00e7on","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref16","article-title":"What matters when building vision-language models?","author":"Lauren\u00e7on","year":"2024","journal-title":"arXiv preprint"},{"key":"ref17","article-title":"Seed-bench: Benchmarking multimodal 11 ms with generative comprehension","author":"Li","year":"2023","journal-title":"arXiv preprint"},{"key":"ref18","article-title":"Llava-onevision: Easy visual task transfer","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref19","article-title":"Cumo: Scaling multimodal 11 m with co-upcycled mixture-of-experts","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3637265"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02520"},{"key":"ref22","volume-title":"Llava-next: Improved reasoning, ocr, and world knowledge","author":"Liu","year":"2024"},{"key":"ref23","article-title":"Visual instruction tuning","volume":"36","author":"Liu","year":"2024","journal-title":"Advances in neural information processing systems"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72658-3_13"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0182"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.9"},{"key":"ref27","article-title":"Im2text: Describing images using 1 million captioned photographs","volume":"24","author":"Ordonez","year":"2011","journal-title":"Advances in neural information processing systems"},{"key":"ref28","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"International conference on machine learning","author":"Radford","year":"2021"},{"key":"ref29","article-title":"Direct preference optimization: Your language model is secretly a reward model","volume":"36","author":"Rafailov","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3406703"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1833"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P18-1238"},{"key":"ref33","article-title":"Megatronlm: Training multi-billion parameter language models using model parallelism","author":"Shoeybi","year":"2019","journal-title":"arXiv preprint"},{"key":"ref34","article-title":"Chameleon: Mixed-modal early-fusion foundation models","year":"2024","journal-title":"arXiv preprint"},{"key":"ref35","article-title":"Gemma 2: Improving open language models at a practical size","author":"Team","year":"2024","journal-title":"arXiv preprint"},{"key":"ref36","article-title":"Cambrian1: A fully open, vision-centric exploration of multimodal 11 ms","author":"Tong","year":"2024","journal-title":"arXiv preprint"},{"key":"ref37","first-page":"9568","article-title":"Eyes wide shut? exploring the visual shortcomings of multimodal 11 ms","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Tong","year":"2024"},{"key":"ref38","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref39","article-title":"Attention is all you need","volume-title":"Advances in Neural Information Processing Systems","author":"Vaswani","year":"2017"},{"key":"ref40","article-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution","author":"Wang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref41","article-title":"Xgen-mm (blip-3): A family of open large multimodal models","author":"Xue","year":"2024","journal-title":"arXiv preprint"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01100"},{"key":"ref43","article-title":"Root mean square layer normalization","volume":"32","author":"Zhang","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref44","article-title":"Internlm-xcomposer-2.5: A versatile large vision language model supporting long-contextual input and output","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2020"}],"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\/11443958.pdf?arnumber=11443958","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T04:59:28Z","timestamp":1777611568000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11443958\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":45,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.02219","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}