[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-657528-105":59,"doc-detail-657528-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","draw-with-thought-unleashing-multimodal-reasoning-for-scientific-diagram-generation","Draw with Thought - Unleashing Multimodal Reasoning for Scientific Diagram Generation","","Scientific diagrams are essential for communicating structured knowledge, yet they are commonly distributed as static raster images that discard symbolic semantics and hinder reuse. Draw with Thought (DwT) is a training-free framework that guides multimodal large language models to reconstruct diagrams into editable mxGraph XML using chain-of-thought style reasoning. The method separates coarse-to-fine planning from structure-aware code generation, enabling controllable, interpretable outputs without fine-tuning. Plot2XML provides benchmark data to evaluate reconstruction fidelity and interpretability.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/draw-with-thought-unleashing-multimodal-reasoning-for-scientific-diagram-generation/657528/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/draw-with-thought-unleashing-multimodal-reasoning-for-scientific-diagram-generation/657528.png","ImageObject",300,407,{"name":92,"@type":93},"Riley West","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-10-26",true,{"@type":102,"interactionType":103,"userInteractionCount":19},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does Draw with Thought address for scientific diagrams?","Question",{"text":112,"@type":113},"It addresses the loss of symbolic semantics when diagrams are distributed as static raster images, making them non-editable and difficult to reuse. DwT reconstructs diagrams into an editable structured representation instead.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does Draw with Thought generate structured diagram code without fine-tuning?",{"text":117,"@type":113},"DwT is training-free and divides the task into two stages: coarse-to-fine planning for perceptual structuring and semantic specification, and structure-aware code generation with format-guided refinement.",{"name":119,"@type":110,"acceptedAnswer":120},"What dataset and evaluation support are provided for this work?",{"text":121,"@type":113},"The paper releases Plot2XML, a benchmark of 247 real-world scientific diagrams with gold-standard XML annotations. Experiments across multiple multimodal models report high-fidelity, semantically aligned, and structurally valid reconstructions with human evaluation confirming accuracy and visual aesthetics.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},657528,1793115203,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":19,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":52,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":143,"read_time":144},1099523885074,"https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc","Draw with Thought: Unleashing Multimodal Reasoning for  \nScientific Diagram Generation  \narXiv :2504 .09479v2 [ cs .AI] 13 Oct 2025  \nZhiqing Cui∗ [zhiqing@nuist.edu.cn](zhiqing@nuist.edu.cn)[ ](zhiqing@nuist.edu.cn)Nanjing University of Information Science & Technology Nanjing, China  \nYanshu Li  \n[yanshu_li1@brown.edu](yanshu_li1@brown.edu)[ ](yanshu_li1@brown.edu)Brown University Providence, America  \nJiahao Yuan∗ [jhyuan.cs@gmail.com](jhyuan.cs@gmail.com)[ ](jhyuan.cs@gmail.com)East China Normal University Shanghai, China  \nChenxu Du  \n[dcx_swjtu@outlook.com](dcx_swjtu@outlook.com)[ ](dcx_swjtu@outlook.com)Southwest Jiaotong University Chengdu, China  \nHanqing Wang  \n[wanghanqing0424@gmail.com](wanghanqing0424@gmail.com)[ ](wanghanqing0424@gmail.com)The Hong Kong University of Science and Technology (Guangzhou) Guangzhou, China  \nZhenglong Ding† [zlding@nuist.edu.cn](zlding@nuist.edu.cn)[ ](zlding@nuist.edu.cn)Nanjing University of Information Science & Technology Nanjing, China  \nFigure 1: Overview of our Draw with Thought framework.  \nAbstract  \nScientific diagrams are vital tools for communicating structured knowledge across disciplines. However, they are often published as static raster images, losing symbolic semantics and limiting reuse. While Multimodal Large Language Models (MLLMs) offera pathway to bridging vision and structure, existing methods lack  \n∗ Both authors contributed equally to this research.†Corresponding Author.  \nPermission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission [and/or a fee. Request permissions from permissions@acm.org](and/or a fee. Request permissions from permissions@acm.org).  \nMM’25, Dublin, Ireland.  \n© 2025 Copyright held by the owner/author(s) . Publication rights licensed to ACM. ACM ISBN 979-8-4007-2035-2/2025/10  \n[https://doi.org/10.1145/3746027.3755731](https://doi.org/10.1145/3746027.3755731)  \nsemantic control and structural interpretability, especially on complex diagrams. We propose Draw with Thought (DwT), a trainingfree framework that guides MLLMs to reconstruct diagrams into editable mxGraph XML code through cognitively inspired Chainof-Thought reasoning. DwT enables interpretable and controllable outputs without model fine-tuning by dividing the task into two stages: Coarse-to-Fine Planning, which handles perceptual structuring and semantic specification, and Structure-Aware Code Generation, enhanced by format-guided refinement. To support evaluation, we release Plot2XML, a benchmark of 247 real-world scientific diagrams with gold-standard XML annotations. Extensive experiments across eight MLLMs show that our approach yields high-fidelity, semantically aligned, and structurally valid reconstructions, with human evaluations confirming strong alignment in both accuracy and visual aesthetics, offering a scalable solution for converting static visuals into structurally valid and renderable representationsand advancing machine understanding of scientific graphics.  \nCCS Concepts  \n• Computing methodologies → Planning with abstraction and generalization.  \nKeywords  \nMultimodal Large Language Models, Chain-of-Thought, Code Generation  \nACM Reference Format:  \nZhiqing Cui, Jiahao Yuan, Hanqing Wang, Yanshu Li, Chenxu Du, and Zhenglong Ding. 2025. Draw with Thought: Unleashing Multimodal Reasoning for Scientific Diagram Generation. In Proceedings of the 33rd ACM International Conference on Multimedia (MM’25), October 27–31, 2025, Dublin, Ireland. ACM, New York, NY, USA, 10 pages. [https://doi.org/10.1145/3746027.3755","cbCaijXgu43w9SKK","https://ap.wps.com/l/cbCaijXgu43w9SKK","pdf",3025795,"English","# Abstract\n## Introduction\n## Scientific diagrams and limitations of raster formats\n## Structured representation recovery as visual perception plus abstraction\n## Related work on target formats (SVG, TikZ, Python, XML)","[{\"question\":\"What problem does Draw with Thought address for scientific diagrams?\",\"answer\":\"It addresses the loss of symbolic semantics when diagrams are distributed as static raster images, making them non-editable and difficult to reuse. DwT reconstructs diagrams into an editable structured representation instead.\"},{\"question\":\"How does Draw with Thought generate structured diagram code without fine-tuning?\",\"answer\":\"DwT is training-free and divides the task into two stages: coarse-to-fine planning for perceptual structuring and semantic specification, and structure-aware code generation with format-guided refinement.\"},{\"question\":\"What dataset and evaluation support are provided for this work?\",\"answer\":\"The paper releases Plot2XML, a benchmark of 247 real-world scientific diagrams with gold-standard XML annotations. Experiments across multiple multimodal models report high-fidelity, semantically aligned, and structurally valid reconstructions with human evaluation confirming accuracy and visual aesthetics.\"}]","Draw with Thought - Unleashing Multimodal Reasoning for Scientific Diagram Generation | PDF",1792982232,25]