[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-657627-105":59,"doc-detail-657627-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","data-dependency-aware-code-generation-from-enhanced-uml-sequence-diagrams","Data Dependency-Aware Code Generation from Enhanced UML Sequence Diagrams","","Large language models excel at generating code from natural language, yet plain descriptions are often ambiguous and fail to capture intricate requirements such as system behaviors, conditional logic, and architectural constraints. Implicit data dependencies in service-oriented architectures remain difficult to infer reliably. UML2Dep addresses this by enhancing UML sequence diagrams and introducing data dependency inference to build explicit dependency graphs before code synthesis, improving accuracy and efficiency in industrial settings.",{"@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/data-dependency-aware-code-generation-from-enhanced-uml-sequence-diagrams/657627/",{"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/data-dependency-aware-code-generation-from-enhanced-uml-sequence-diagrams/657627.png","ImageObject",300,407,{"name":92,"@type":93},"asba","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-10-26",true,{"@type":102,"interactionType":103,"userInteractionCount":24},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does UML2Dep target in LLM-based code generation?","Question",{"text":112,"@type":113},"It targets the ambiguity of natural-language requirements and, critically, the difficulty of correctly inferring implicit data dependencies in service-oriented architectures from UML sequence diagrams.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does UML2Dep improve sequence diagram specifications?",{"text":117,"@type":113},"It introduces an enhanced UML sequence diagram that integrates decision tables and API specifications, formalizing structural relationships and business logic flows to reduce linguistic ambiguity.",{"name":119,"@type":110,"acceptedAnswer":120},"What is the role of the DDI task in UML2Dep?",{"text":121,"@type":113},"DDI constructs an explicit data dependency graph before code synthesis, and it is formulated as constrained mathematical reasoning with prompting strategies plus static parsing and dependency pruning to improve reliability.","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},657627,1793052201,{"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":24,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145},3985747872567,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Data Dependency-Aware Code Generation from Enhanced UML Sequence Diagrams  \nWenxin Mao 1 ,†,‡, Zhitao Wang 1 ,†, Long Wang 1 ,∗ , Sirong Chen 1 ,‡, Cuiyun Gao2 ,∗ ,  \nLuyang Cao 1 , Ziming Liu 1 , Qiming Zhang 1 , Jun Zhou 1 , Zhi Jin3  \n1 WeChat Pay, Tencent Inc, Shenzhen, China  \n2 The Chinese University of Hong Kong, Hong Kong, China  \n3 Wuhan University, Wuhan, China  \n[mao23x@qq.com](mao23x@qq.com), {zhitaowang, oliverlwang, [sirongchen](sirongchen}@tencent.com)[}](sirongchen}@tencent.com)[@tencent.com](sirongchen}@tencent.com), [cuiyungao@outlook.com](cuiyungao@outlook.com),  \n{lucascao, simonzmliu, archrzhang, [anderszhou](anderszhou}@tencent.com)[}](anderszhou}@tencent.com)[@tencent.com](anderszhou}@tencent.com), [zhijin@pku.edu.cn](zhijin@pku.edu.cn)  \narXiv :2508 .03379v 3 [ cs .AI ] 5 Nov 2025  \nAbstract—Large language models (LLMs) excel at generating code from natural language (NL) descriptions. However, the plain textual descriptions are inherently ambiguous and often fail to capture complex requirements like intricate system behaviors, conditional logic, and architectural constraints; implicit data dependencies in service-oriented architectures are difficult to infer and handle correctly.  \nTo bridge this gap, we propose a novel step-by-step code generation framework named UML2Dep by leveraging unambiguous formal specifications of complex requirements. First, we introduce an enhanced Unified Modeling Language (UML) sequence diagram tailored for service-oriented architectures. This diagram extends traditional visual syntax by integrating decision tables and API specifications, explicitly formalizing structural relationships and business logic flows in service interactions to rigorously eliminate linguistic ambiguity. Second, recognizing the critical role of data flow, we introduce a dedicated data dependency inference (DDI) task. DDI systematically constructs an explicit data dependency graph prior to actual code synthesis. To ensure reliability, we formalize DDI as a constrained mathematical reasoning task through novel prompting strategies, aligning with LLMs’ excellent mathematical strengths. Additional static parsing and dependency pruning further reduce context complexity and cognitive load associated with intricate specifications, thereby enhancing reasoning accuracy and efficiency.  \nExperimental results on our in-house industrial datasets demonstrate the effectiveness of the proposed framework. Specifically, our framework achieves strong performance, with 89.97% recall, 95.06% precision, and 92.33% F1 score on the DDI task. Furthermore, the integration of UML2Dep into the code generation pipeline also improves practical deployment, increasing compilation pass rate by 8.83% and unit test pass rate by 11.66% .  \nIndex Terms—Data Dependency Inference, UML, sequence diagram, code generation  \nI. INTRODUCTION  \nLarge language models (LLMs) have demonstrated remarkable capabilities in automated code synthesis from natural language (NL) descriptions, transforming how developers approach programming tasks [1]–[6] . However, when scaling from simple demonstrations to production-grade software sys  \ntems, a fundamental challenge persists: the inherent impre-†These authors contributed equally to this work.  \n‡The work was done during an internship at Tencent.  \n∗ These authors are the corresponding authors.  \ncision of natural language becomes increasingly problematic for capturing the detailed specifications required by complex applications. Natural language descriptions often lack the structural clarity needed to unambiguously define sophisticated system interactions, execution flows, and data relationships [7],[8] .  \nContemporary software engineering practice increasingly adopts Unified Modeling Language (UML) sequence diagrams as design blueprints [9] . These diagrams provide formal visual syntax that explicitly captures control flows and service interactions with a obvious advancement over NL specifications, a","cbCaiqgZ7NtxK83b","https://ap.wps.com/l/cbCaiqgZ7NtxK83b","pdf",2478229,11,"English","# I. INTRODUCTION\n## Data Dependency Inference (DDI)\n## UML2Dep Framework","[{\"question\":\"What problem does UML2Dep target in LLM-based code generation?\",\"answer\":\"It targets the ambiguity of natural-language requirements and, critically, the difficulty of correctly inferring implicit data dependencies in service-oriented architectures from UML sequence diagrams.\"},{\"question\":\"How does UML2Dep improve sequence diagram specifications?\",\"answer\":\"It introduces an enhanced UML sequence diagram that integrates decision tables and API specifications, formalizing structural relationships and business logic flows to reduce linguistic ambiguity.\"},{\"question\":\"What is the role of the DDI task in UML2Dep?\",\"answer\":\"DDI constructs an explicit data dependency graph before code synthesis, and it is formulated as constrained mathematical reasoning with prompting strategies plus static parsing and dependency pruning to improve reliability.\"}]","Data Dependency-Aware Code Generation from Enhanced UML Sequence Diagrams | PDF",1792985439,28]