[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-272999-105":59,"doc-detail-272999-en":131},{"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":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","towards-realistic-project-level-code-generation-via-multi-agent-collaboration-and-semantic-architecture-modeling","Towards Realistic Project-Level Code Generation via Multi-Agent Collaboration and Semantic Architecture Modeling","","Large Language Models (LLMs) are advancing automated code generation, yet real-world software engineering demands project-level outputs that remain coherent across large inputs and architectural constraints. Existing work is limited by unrealistic datasets and evaluation metrics, semantic mismatch between requirements and implementation, and weak handling of hierarchical dependencies and quality control. To address this, the study introduces CodeProjectEval and proposes ProjectGen with a multi-agent pipeline plus the Semantic Software Architecture Tree (SSAT), enabling semantic bridging and improved test-based performance.",{"@graph":69,"@context":123},[70,84,106],{"@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/towards-realistic-project-level-code-generation-via-multi-agent-collaboration-and-semantic-architecture-modeling/272999/",{"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/towards-realistic-project-level-code-generation-via-multi-agent-collaboration-and-semantic-architecture-modeling/272999.png","ImageObject",300,407,{"name":92,"@type":93},"Kyle","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-09-15",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",18,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What challenge does project-level code generation face in practice?","Question",{"text":113,"@type":114},"It requires generating complete software projects from complex requirements while handling larger input scales, architectural reasoning, and quality across hierarchical dependencies.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does CodeProjectEval support automatic evaluation?",{"text":118,"@type":114},"It is built from real-world repositories and includes documentation and executable test cases for automatic, test-based assessment.",{"name":120,"@type":111,"acceptedAnswer":121},"What is the role of the Semantic Software Architecture Tree (SSAT) in ProjectGen?",{"text":122,"@type":114},"SSAT provides a semantically rich structure that bridges user requirements to source code implementation during the generation process.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},272999,1789468902,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":130,"read_time":145,"preview_page_count":4},3985741905716,"https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d","arXiv :2511 .03404v 1 [ cs . SE] 5 Nov 2025  \nTowards Realistic Project-Level Code Generation via Multi-Agent Collaboration and Semantic Architecture Modeling  \nQIANHUI ZHAO, Beihang University, China LI ZHANG, Beihang University, China  \nFANG LIU∗ , Beihang University, China JUNHANG CHENG, Beihang University, China CHENGRU WU, Beihang University, China JUNCHEN AI, Beihang University, China QIAOYUANHE MENG, Beihang University, China LICHEN ZHANG, Independent, China  \nXIAOLI LIAN, Beihang University, China  \nSHUBIN SONG, Software IDE Innovation Lab, Huawei Central Software Institute, China YUANPING GUO, Software IDE Innovation Lab, Huawei Central Software Institute, China  \nIn recent years, Large Language Models (LLMs) have achieved remarkable progress in automated code generation. In real-world software engineering, the growing demand for rapid iteration and continuous delivery underscores the importance of project-level code generation, where LLMs are expected to generate complete software projects directly from complex user requirements. This task, however, presents greater challenges due to larger input scales, multi-modal information, and the need for holistic architectural reasoning. Although existing studies have made initial explorations, they still face key limitations, including unrealistic datasets and unreliable evaluation metrics that fail to reflect real-world complexity, the semantic gap between human-written requirements and machine-interpretable structures, and difficulties in managing hierarchical dependencies and maintaining quality throughout the generation process. To address these limitations, we first introduce CodeProjectEval, a project-level code generation dataset built from 18 real-world repositories with 12.7 files and 2,388.6 lines of code per task on average, supplemented with documentation and executable test cases for automatic evaluation. We further propose ProjectGen, a multi-agent framework that decomposes projects into architecture design, skeleton generation, and code filling stages with iterative refinement and memory-based context management. Within this framework, we introduce the Semantic Software Architecture Tree (SSAT), a structured and semantically rich representation that effectively bridges user requirements and source code implementation. Experiments show that ProjectGen  \n∗ Corresponding author.  \nAuthors’ Contact Information: Qianhui Zhao, [zhaoqianhui@buaa.edu.cn](zhaoqianhui@buaa.edu.cn), Beihang University, Beijing, China; Li Zhang, Beihang University, Beijing,  \nChina, [lily@buaa.edu.cn](lily@buaa.edu.cn); Fang Liu, Beihang University, Beijing, China, [fangliu@buaa.edu.cn](fangliu@buaa.edu.cn); Junhang Cheng, Beihang University, Beijing, China,  \n[chengjunhang7@gmail.com](chengjunhang7@gmail.com); Chengru Wu, Beihang University, Beijing, China, [chengru_wu@buaa.edu.cn](chengru_wu@buaa.edu.cn); Junchen Ai, Beihang University, Beijing, China, [junchen_ai@buaa.edu.cn](junchen_ai@buaa.edu.cn); Qiaoyuanhe Meng, Beihang University, Beijing, China, [mengqiaoyuanhe@buaa.edu.cn](mengqiaoyuanhe@buaa.edu.cn); Lichen Zhang, Independent, Beijing,  \nChina, [lczhang9653@gmail.com](lczhang9653@gmail.com); Xiaoli Lian, Beihang University, Beijing, China, [lianxiaoli@buaa.edu.cn](lianxiaoli@buaa.edu.cn); Shubin Song, Software IDE Innovation Lab,  \nHuawei Central Software Institute, China, [songshubin2@huawei.com](songshubin2@huawei.com); Yuanping Guo, Software IDE Innovation Lab, Huawei Central Software Institute, China, [guoyuanping1@huawei.com](guoyuanping1@huawei.com).  \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 otherwis","cbCaifZ079jz5xjs","https://ap.wps.com/l/cbCaifZ079jz5xjs","pdf",1712811,26,"English","# Introduction\n## Problem background and limitations\n## Dataset: CodeProjectEval\n## Framework: ProjectGen\n## Semantic representation: SSAT\n## Experiments and results","[{\"question\":\"What challenge does project-level code generation face in practice?\",\"answer\":\"It requires generating complete software projects from complex requirements while handling larger input scales, architectural reasoning, and quality across hierarchical dependencies.\"},{\"question\":\"How does CodeProjectEval support automatic evaluation?\",\"answer\":\"It is built from real-world repositories and includes documentation and executable test cases for automatic, test-based assessment.\"},{\"question\":\"What is the role of the Semantic Software Architecture Tree (SSAT) in ProjectGen?\",\"answer\":\"SSAT provides a semantically rich structure that bridges user requirements to source code implementation during the generation process.\"}]","Towards Realistic Project-Level Code Generation via Multi-Agent Collaboration and Semantic Architecture Modeling | PDF",66]