{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T09:10:35Z","timestamp":1780477835413,"version":"3.54.1"},"reference-count":94,"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\/iccv51701.2025.02040","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"21972-21983","source":"Crossref","is-referenced-by-count":4,"title":["SmolDocling: An Ultra-Compact Vision-Language Model for End-To-End Multi-Modal Document Conversion"],"prefix":"10.1109","author":[{"given":"Ahmed","family":"Nassar","sequence":"first","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Matteo","family":"Omenetti","sequence":"additional","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maksym","family":"Lysak","sequence":"additional","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nikolaos","family":"Livathinos","sequence":"additional","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christoph","family":"Auer","sequence":"additional","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lucas","family":"Morin","sequence":"additional","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rafael Teixeira","family":"de Lima","sequence":"additional","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yusik Kim A. Said","family":"Gurbuz","sequence":"additional","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michele","family":"Dolfi","sequence":"additional","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peter W. J.","family":"Staar","sequence":"additional","affiliation":[{"name":"IBM Research"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","first-page":"2008","volume-title":"Grobid."},{"key":"ref2","first-page":"2","author":"Ben Allal","year":"2025","journal-title":"SmolLM2: When Smol Goes Big - Data-Centric Training of a Small Language Model"},{"key":"ref3","volume-title":"Amazon Web Services. Amazon Textract","year":"2024"},{"key":"ref4","article-title":"Palm 2 technical report","volume-title":"arXiv preprint","author":"Anil","year":"2023"},{"key":"ref5","volume-title":"Claude-3.5.","year":"2024"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CLOUD55607.2022.00060"},{"key":"ref7","first-page":"5","article-title":"Docling Technical Report","volume-title":"arXiv preprint","author":"Auer","year":"2024"},{"key":"ref8","first-page":"2","article-title":"Qwen technical report","volume-title":"arXiv preprint","author":"Bai","year":"2023"},{"key":"ref9","first-page":"2","volume-title":"Qwen2.5-VL Technical Report","author":"Bai","year":"2025"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-25069-9_16"},{"key":"ref11","year":"2024","journal-title":"Lukas Blecher. LaTeX-OCR"},{"key":"ref12","first-page":"3","article-title":"Nougat: Neural optical understanding for academic documents","author":"Blecher","year":"2023","journal-title":"arXiv preprint"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3664647.3681167"},{"issue":"70","key":"ref14","first-page":"1","article-title":"Scaling instructionfinetuned language models","volume":"25","author":"Won Chung","year":"2024","journal-title":"Journal of Machine Learning Research"},{"key":"ref15","year":"2024","journal-title":"Pygments: Python syntax highlighter"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-25069-9_19"},{"key":"ref17","first-page":"980","article-title":"Rush","volume-title":"Image-to-markup generation with coarse-tofine attention. In International Conference on Machine Learning","author":"Deng"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-41734-4_10"},{"key":"ref19","article-title":"InternLM-X Composer 2: Mastering Free-form Text-Image Composition and Comprehension in VisionLanguage Large Model","author":"Dong","year":"2024","journal-title":"arXiv preprint"},{"key":"ref20","article-title":"InternLM-X Composer 24KHD: A Pioneering Large Vision-Language Model Handling Resolutions from 336 Pixels to 4 K HD","author":"Dong","year":"2024","journal-title":"arXiv preprint"},{"key":"ref21","volume-title":"Gregory Eritsyan. Im2LaTeX 230K.","year":"2025"},{"key":"ref22","year":"2024","journal-title":"Hugging Face. latex-formulas Dataset"},{"key":"ref23","first-page":"3","volume-title":"Mathwriting: A dataset for handwritten mathematical expression recognition","author":"Gervais","year":"2024"},{"key":"ref24","volume-title":"Google Cloud. Document AI.","year":"2024"},{"key":"ref25","first-page":"2","article-title":"Team","author":"Granite","year":"2024","journal-title":"Granite 3.0 Language Models"},{"key":"ref26","first-page":"3","article-title":"Chartllama: A multimodal 11 m for chart understanding and generation","author":"Han","year":"2023","journal-title":"arXiv preprint"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-emnlp.175"},{"key":"ref28","article-title":"mplug-docow12: High-resolution compressing for ocrfree multi-page document understanding","author":"Hu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3548112"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/ICDARW.2019.10029"},{"key":"ref31","article-title":"Mixtral of experts","author":"Jiang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref32","first-page":"3","article-title":"Figureqa: An annotated figure dataset for visual reasoning","author":"Ebrahimi Kahou","year":"2017","journal-title":"arXiv preprint"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.277"},{"issue":"15","key":"ref34","first-page":"2","article-title":"Donut: Document understanding transformer without ocr","volume":"7","author":"Kim","year":"2021","journal-title":"arXiv preprint"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"ref36","author":"Kocetkov","year":"2022","journal-title":"The Stack: 3 TB of permissively licensed source code. Preprint"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref38","first-page":"2","author":"Lauren\u00e7on","year":"2024","journal-title":"Building and better understanding visionlanguage models: insights and future directions"},{"key":"ref39","first-page":"3","author":"Lauren\u00e7on","year":"2024","journal-title":"What matters when building vision-language models?"},{"key":"ref40","first-page":"18893","article-title":"Pix2struct: Screenshot parsing as pretraining for visual language understanding","volume-title":"In International Conference on Machine Learning","author":"Lee"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/1148170.1148307"},{"key":"ref42","first-page":"2","article-title":"Llava-onevision: Easy visual task transfer","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref43","article-title":"Llava-next-interleave: Tackling multi-image, video, and 3d in large multimodal models","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref44","first-page":"1973019742","article-title":"Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","volume-title":"In International conference on machine learning","author":"Li"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1007\/s11831-019-09315-1"},{"key":"ref46","article-title":"Deplot: One-shot visual language reasoning by plot-to-table translation","author":"Liu","year":"2022","journal-title":"arXiv preprint"},{"key":"ref47","first-page":"3","article-title":"Matcha: Enhancing visual language pretraining with math reasoning and chart derendering","author":"Liu","year":"2022","journal-title":"arXiv preprint"},{"key":"ref48","first-page":"2","author":"Liu","year":"2023","journal-title":"Improved Baselines with Visual Instruction Tuning"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref50","first-page":"2","author":"Livathinos","year":"2025","journal-title":"Docling: An Efficient Open-Source Toolkit for AI-driven Document Conversion"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.5220\/0011682600003411"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-41679-8_3"},{"key":"ref53","first-page":"1","author":"Marafioti","year":"2025","journal-title":"SmolVLM: Redefining small and efficient multimodal models"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.findings-acl.177"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.906"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/WACV48630.2021.00225"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.463"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/WACV45572.2020.9093523"},{"key":"ref59","volume-title":"Microsoft Azure. AI Document Intelligence.","year":"2024"},{"key":"ref60","volume-title":"OpenAI. Hello GPT-4O.","year":"2024"},{"issue":"5","key":"ref61","volume-title":"R OpenAI. Gpt-4 technical report.","volume":"2","year":"2023"},{"key":"ref62","first-page":"2","volume-title":"Marker: Convert PDF to Markdown Quickly with High Accuracy.","author":"Paruchuri","year":"2024"},{"key":"ref63","volume-title":"Texify.","author":"Paruchuri","year":"2024"},{"key":"ref64","article-title":"Kosmos-2: Grounding multimodal large language models to the world","author":"Peng","year":"2023","journal-title":"arXiv preprint"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1145\/3534678.3539043"},{"key":"ref66","author":"Puri","year":"2021","journal-title":"CodeNet: A Large-Scale AI for Code Dataset for Learning a Diversity of Coding Tasks"},{"issue":"140","key":"ref67","first-page":"1","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"Journal of machine learning research"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3404834"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00459"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01845"},{"key":"ref71","article-title":"Gemini: a family of highly capable multimodal models","author":"Team","year":"2023","journal-title":"arXiv preprint"},{"key":"ref72","author":"Vision Team","year":"2025","journal-title":"Granite Vision: a lightweight, open-source multimodal model for enterprise Intelligence"},{"key":"ref73","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-41682-8_22"},{"key":"ref75","volume-title":"Unstructured.io: OpenSource Pre-Processing Tools for Unstructured Data.","year":"2024"},{"key":"ref76","volume-title":"Vicuna: An open-source chatbot impressing GPT4 with 90 %* ChatGPT quality.","year":"2023"},{"key":"ref77","first-page":"2","author":"Wang","year":"2024","journal-title":"MinerU: An Open-Source Solution for Precise Document Content Extraction"},{"key":"ref78","first-page":"5","article-title":"WordScape: a Pipeline to extract multilingual, visually rich Documents with Layout Annotations from Web Crawl Data","volume":"36","author":"Weber","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref79","first-page":"3","article-title":"Theory: Towards OCR-2.0 via a Unified End-to-end Model","author":"Wei","year":"2024","journal-title":"arXiv preprint"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1800"},{"key":"ref81","volume-title":"Pandas: powerful Python data analysis toolkit, Release 1.4.4.","author":"McKinney","year":"2022"},{"key":"ref82","first-page":"3","article-title":"StructChart: Perception, Structuring, Reasoning for Visual Chart Understanding","author":"Xia","year":"2023","journal-title":"arXiv preprint"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/tip.2025.3607618"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3403172"},{"key":"ref85","article-title":"Layoutlmv2: Multi-modal pre-training for visually-rich document understanding","author":"Xu","year":"2020","journal-title":"arXiv preprint"},{"key":"ref86","doi-asserted-by":"publisher","DOI":"10.1109\/ICPR48806.2021.9412913"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.187"},{"key":"ref88","article-title":"InternLM-X Composer: A Vision-Language Large Model for Advanced Text-image Comprehension and Composition","author":"Zhang","year":"2023","journal-title":"arXiv preprint"},{"key":"ref89","article-title":"InternLM-X Composer- 2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2023.3347693"},{"key":"ref91","first-page":"19","volume":"3","author":"Zhang","year":"2023","journal-title":"Opt: Open pretrained transformer language models"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.1109\/WACV48630.2021.00074"},{"key":"ref93","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58589-1_34"},{"key":"ref94","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\/11446172.pdf?arnumber=11446172","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T06:16:29Z","timestamp":1777529789000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11446172\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":94,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.02040","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}