[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86479-en":3,"doc-seo-86479-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86479,687197207639,"Asher","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","ChartSync: A Benchmark for Visuo-Logical Cascading Chart Editing","ChartSync introduces a structured benchmark for Visuo-Logical Cascading Editing (VLCE) in statistical charts, where data changes must trigger geometrically synchronized updates rather than isolated text replacement. The benchmark builds 870 triplets across nine chart categories and four task types, including 235 geometry-coupled VLCE instances focused on cascading text-to-geometry synchronization. A two-tier evaluation combines objective visual metrics and a vision-language model judge to measure fidelity and reasoning. Tests on 14 editing models reveal large synchronization gaps with only two proprietary methods showing emerging VLCE capability.","ChartSync: A Benchmark for Visuo-Logical Cascading Chart Editing  \nJiakang Yu1,2,*,†, Yixuan Chai2,* , Tianci Wang3 , Rihui Jin4 , Guangkai Xu5 , Hongtao Deng1,‡, Xun Zhu1 , Wang Gao1,‡, Xinrun Guo2 , Haipang Wu2  \n1Jianghan University 2HiThink Research  \n3University of Science and Technology of China  \n4 Southeast University 5Zhejiang University  \n[hongtaodeng@jhun.edu.cn](hongtaodeng@jhun.edu.cn) , [gaow@jhun.edu.cn](gaow@jhun.edu.cn)  \narXiv :2607 . 10301v1 [ cs .CV] 11 Jul 2026  \nAbstract  \nGenerative image editing models struggle with structured statistical charts when data modifications require geometric synchronization. We formalize this task as Visuo-Logical Cascading Editing (VLCE) . However, existing methods remain confined to localized text substitutions and struggle with dependency-aware cascading updates. To systematically evaluate this capability, we introduce ChartSync, an expert-validated benchmark constructed via a programmatic rendering pipeline that guarantees deterministic visuo-logical coupling for the ground truth. ChartSync comprises 870 triplets across 9 chart categories and 4 task types, including 235 geometry-coupled VLCE instances that specifically test cascading text-to-geometry synchronization. We further evaluate these instances via a two-tier framework combining objective visual metrics with a vision-language model judge paradigm to assess low-level fidelity alongside multimodal comprehension and reasoning. Evaluating 14 image editing models and one code-mediated pipeline reveals a nuanced capability gap: most open-source models suffer severe drops in geometric synchronization, while only two frontier proprietary models show emerging VLCE capability, with their residual errors mainly involving semantic isolation and background corruption. Our detailed error analysis deconstructs these failure paradigms to identify core meta-abilities for guiding future multimodal architectures.  \nThe ChartSync dataset and code are publicly released at [https://github.com/kaka-yjk/](https://github.com/kaka-yjk/)[ ](https://github.com/kaka-yjk/)ChartSyncCodebase.  \n1 Introduction  \nWhile instruction-based image editing has advanced rapidly in natural scenes (Brooks et al., 2023 ; Zhang et al., 2023 ; Pan et al., 2025 ; Pathiraja  \n*  \nEqual contribution.  \n†Work done during an internship at HiThink Research.‡Corresponding authors.  \n\n|  |\n| --- |\n|  |\n\nFigure 1: An illustration of the VLCE task. While traditional instruction-based editing merely substitutes text labels, our VLCE requirement demands a synchronous spatial alignment.  \net al., 2025 ; Ma et al., 2024), extending this to statistical charts remains challenging due to precise text-geometry synchronization (Gui et al., 2025) .  \nRecent chart editing works often adopt a codebased paradigm, using intermediate scripts to avoid direct pixel generation (Zhao et al., 2025 ; Yang et al., 2025b) . However, this paradigm assumes source-code availability for flattened chart images and can suffer from information loss during reverse engineering (Yang et al., 2025a ; Tang et al., 2025) . This motivates direct pixel-space chart editing for real-world flattened chart images.  \nNevertheless, direct chart manipulation introduces challenges spanning multimodal comprehension, reasoning, and generation. Even simple text modifications demand precise layout alignment to prevent visual corruption (Gui et al., 2025) . Furthermore, unlike natural scenes that allow geometric flexibility, chart geometries are governed by underlying data. As argued in (Li et al., 2025), chart editing is a structured transformation problem rather than generic image manipulation. We formal-  \n\n| Benchmark | Domain | Input at Inference | Output / Target | Core Evaluated Ability | Value-to-Geometry Sync. | Sync. Metric |\n| --- | --- | --- | --- | --- | --- | --- |\n| OmniText (Gunawan et al., 2025) | Visual text | Image + mask + text/style | Image | Text content/style manipulation | No | No |\n| TextEditB","cbCaiohSNuxT0bal","https://ap.wps.com/l/cbCaiohSNuxT0bal","pdf",1700078,4,1,22,"English","en",105,"# Abstract\n# Introduction\n## Related Chart Editing Benchmarks","[{\"question\":\"What problem does VLCE address in statistical chart editing?\",\"answer\":\"VLCE requires synchronizing text modifications with corresponding geometric adjustments driven by the underlying data. This prevents visual corruption that occurs when labels change without layout updates.\"},{\"question\":\"How is the ChartSync benchmark constructed and what does it contain?\",\"answer\":\"ChartSync uses a programmatic rendering pipeline to guarantee deterministic visuo-logical coupling. It includes 870 triplets across 9 chart categories and 4 task types, with 235 geometry-coupled VLCE instances.\"},{\"question\":\"How are VLCE capabilities evaluated in ChartSync?\",\"answer\":\"Evaluation uses a two-tier framework: objective visual metrics for low-level fidelity and a vision-language model judge paradigm for multimodal comprehension and reasoning.\"}]",1784212025,55,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"chartsync-a-benchmark-for-visuo-logical-cascading-chart-editing","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/chartsync-a-benchmark-for-visuo-logical-cascading-chart-editing/86479/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-28","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does VLCE address in statistical chart editing?","Question",{"text":75,"@type":76},"VLCE requires synchronizing text modifications with corresponding geometric adjustments driven by the underlying data. This prevents visual corruption that occurs when labels change without layout updates.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the ChartSync benchmark constructed and what does it contain?",{"text":80,"@type":76},"ChartSync uses a programmatic rendering pipeline to guarantee deterministic visuo-logical coupling. It includes 870 triplets across 9 chart categories and 4 task types, with 235 geometry-coupled VLCE instances.",{"name":82,"@type":73,"acceptedAnswer":83},"How are VLCE capabilities evaluated in ChartSync?",{"text":84,"@type":76},"Evaluation uses a two-tier framework: objective visual metrics for low-level fidelity and a vision-language model judge paradigm for multimodal comprehension and reasoning.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]