{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T16:00:11Z","timestamp":1778256011225,"version":"3.51.4"},"reference-count":84,"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\/iccv51701.2025.01930","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"20758-20769","source":"Crossref","is-referenced-by-count":2,"title":["The Scalability of Simplicity: Empirical Analysis of Vision-Language Learning with a Single Transformer"],"prefix":"10.1109","author":[{"given":"Weixian","family":"Lei","sequence":"first","affiliation":[{"name":"Bytedance Seed"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiacong","family":"Wang","sequence":"additional","affiliation":[{"name":"Bytedance Seed"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haochen","family":"Wang","sequence":"additional","affiliation":[{"name":"Bytedance Seed"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiangtai","family":"Li","sequence":"additional","affiliation":[{"name":"Bytedance Seed"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jun Hao","family":"Liew","sequence":"additional","affiliation":[{"name":"Bytedance Seed"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiashi","family":"Feng","sequence":"additional","affiliation":[{"name":"Bytedance Seed"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zilong","family":"Huang","sequence":"additional","affiliation":[{"name":"Bytedance Seed"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1723"},{"key":"ref2","article-title":"The claude 3 model family: Opus, sonnet, haiku","year":"2024","journal-title":"Claude-3 Model Card"},{"key":"ref3","article-title":"Qwen-vl: A frontier large vision-language model with versatile abilities","author":"Bai","year":"2023","journal-title":"arXiv preprint"},{"key":"ref4","article-title":"Beit: Bert pre-training of image transformers","author":"Bao","year":"2021","journal-title":"arXiv preprint"},{"key":"ref5","author":"Bavishi","year":"2023","journal-title":"Introducing our multimodal models"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr46437.2021.00356"},{"key":"ref7","article-title":"Minigpt-v2: large language model as a unified interface for vision-language multi-task learning","author":"Chen","year":"2023","journal-title":"arXiv preprint"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01231"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0850"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73004-7_2"},{"key":"ref11","article-title":"A single transformer for scalable vision-language modeling","author":"Chen","year":"2024","journal-title":"Transactions on Machine Learning Research"},{"key":"ref12","article-title":"Vision transformer adapter for dense predictions","author":"Chen","year":"2022","journal-title":"arXiv preprint"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4231-5"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02283"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00276"},{"key":"ref16","author":"Chiang","year":"2023","journal-title":"Vicuna: An open-source chatbot impressing gpt-4 with 90 %* chatgpt quality"},{"key":"ref17","article-title":"MMSegmentation: Openmmlab semantic segmentation toolbox and benchmark","volume-title":"MMSegmentation Contributors","year":"2020"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2142"},{"key":"ref19","first-page":"7480","article-title":"Scaling vision transformers to 22 billion parameters","volume-title":"In International Conference on Machine Learning","author":"Dehghani","year":"2023"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.52202\/079017-1665"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.01953"},{"key":"ref23","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","volume-title":"International Conference on Learning Representations","author":"Dosovitskiy","year":"2021"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3664647.3685520"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01855"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1179"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.670"},{"key":"ref28","article-title":"Infinity-mm: Scaling multimodal performance with large-scale and high-quality instruction data","author":"Gu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01363"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00975"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01553"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.01354"},{"key":"ref33","article-title":"Tag2text: Guiding vision-language model via image tagging","author":"Huang","year":"2023","journal-title":"arXiv preprint"},{"key":"ref34","article-title":"Classification done right for vision-language pretraining","author":"Huang","year":"2024","journal-title":"NeurIPS"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00686"},{"key":"ref36","first-page":"4904","article-title":"Scaling up visual and vision-language representation learning with noisy text supervision","volume-title":"International conference on machine learning","author":"Jia","year":"2021"},{"key":"ref37","article-title":"Mistral 7b","author":"Jiang","year":"2023","journal-title":"arXiv preprint"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_15"},{"key":"ref39","article-title":"Seed-bench: Benchmarking multimodal 11 ms with generative comprehension","author":"Li","year":"2023","journal-title":"arXiv preprint"},{"key":"ref40","article-title":"Llava-onevision: Easy visual task transfer","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2132"},{"key":"ref42","article-title":"What if we recaption billions of web images with llama-3?","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.20"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02240"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1145\/3672758.3672824"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref47","author":"Liu","year":"2024","journal-title":"Llava-next: Improved reasoning, ocr, and world knowledge"},{"key":"ref48","article-title":"Visual instruction tuning","volume":"36","author":"Liu","year":"2024","journal-title":"Advances in neural information processing systems"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72658-3_13"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4235-6"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0182"},{"key":"ref52","article-title":"Mathvista: Evaluating mathematical reasoning of foundation models in visual contexts","author":"Lu","year":"2023","journal-title":"arXiv preprint"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.02324"},{"key":"ref54","article-title":"Catlip: Cliplevel visual recognition accuracy with 2.7 \u00d7 faster pretraining on web-scale image-text data","author":"Mehta","year":"2024","journal-title":"arXiv preprint"},{"key":"ref55","article-title":"Meta","year":"2024","journal-title":"Introducing meta llama 3: The most capable openly available 11 m to date"},{"key":"ref56","article-title":"Introducing chatgpt","year":"2021","journal-title":"OpenAI Blog"},{"key":"ref57","article-title":"Gpt-4 technical report","year":"2023","journal-title":"OpenAI"},{"key":"ref58","article-title":"Gpt-4v(ision) system card","year":"2023","journal-title":"OpenAI"},{"key":"ref59","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"In International conference on machine learning","author":"Radford","year":"2021"},{"key":"ref60","article-title":"LAION-5B: An open large-scale dataset for training next generation image-text models","author":"Schuhmann","year":"2022","journal-title":"In NeurIPS"},{"key":"ref61","article-title":"Megatronlm: Training multi-billion parameter language models using model parallelism","author":"Shoeybi","year":"2019","journal-title":"arXiv preprint"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00851"},{"key":"ref63","volume-title":"SlimPajama: A 627B token cleaned and deduplicated version of RedPajama","author":"Soboleva","year":"2023"},{"key":"ref64","article-title":"Chameleon: Mixed-modal early-fusion foundation models","year":"2024","journal-title":"arXiv preprint"},{"key":"ref65","article-title":"Gemini: a family of highly capable multimodal models","author":"Team","year":"2023","journal-title":"arXiv preprint"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2771"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2029"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2029"},{"key":"ref69","article-title":"Reconstructive visual instruction tuning","volume-title":"International Conference on Learning Representations","author":"Wang","year":"2025"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.265"},{"key":"ref71","author":"Wang","year":"2024","journal-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution"},{"key":"ref72","article-title":"Diffusion feedback helps clip see better","author":"Wang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref73","article-title":"Emu3: Next-token prediction is all you need","author":"Wang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref74","article-title":"SimVLM: Simple visual language model pretraining with weak supervision","volume-title":"International Conference on Learning Representations","author":"Wang","year":"2022"},{"key":"ref75","volume-title":"X.a. Grok-1.5 vision preview","year":"2024"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01228-1_26"},{"key":"ref77","article-title":"Coca: Contrastive captioners are image-text foundation models","author":"Yu","year":"2022","journal-title":"Transactions on Machine Learning Research"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01330"},{"key":"ref79","article-title":"Mm -vet: Evaluating large multimodal models for integrated capabilities","author":"Yu","year":"2023","journal-title":"arXiv preprint"},{"key":"ref80","article-title":"When and why visionlanguage models behave like bags-of-words, and what to do about it?","author":"Yuksekgonul","year":"2022","journal-title":"arXiv preprint"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01100"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW63382.2024.00179"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.544"},{"key":"ref84","article-title":"Minigpt-4: Enhancing vision-language understanding with advanced large language models","author":"Zhu","year":"2023","journal-title":"arXiv preprint"}],"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\/11445583.pdf?arnumber=11445583","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:18:02Z","timestamp":1777612682000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11445583\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":84,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01930","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}