[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83394-en":3,"doc-seo-83394-105":30,"detail-sidebar-cat-0-en-105":90},{"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},83394,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"Research & Report","DocMaster: A Hierarchical Structure-Aware System for Document Analysis","DocMaster is a hierarchical structure-aware system for analyzing complex documents using large language models. It addresses limitations of existing approaches that flatten documents into plain-text chunks, losing structural relationships across sections, tables, figures, and equations. DocMaster parses documents into hierarchical trees, builds structure-aware semantic indices with LLM-guided constrained clustering and cross-section hyper-edges, then applies tri-modal retrieval to filter relevant documents and uses grounded context for retrieval-augmented question answering.","DocMaster: A Hierarchical Structure-Aware System for  \nDocument Analysis  \nZiqi Chen  \nThe Chinese University of Hong Kong, Shenzhen Shenzhen, China[ziqichen1@link.cuhk.edu.cn](ziqichen1@link.cuhk.edu.cn)  \nQuanqing Xu  \nOceanBase, AntGroup Hangzhou, China [xuquanqing.xqq@oceanbase.com](xuquanqing.xqq@oceanbase.com)  \nYingli Zhou  \nThe Chinese University of Hong Kong, Shenzhen Shenzhen, China [yinglizhou@link.cuhk.edu.cn](yinglizhou@link.cuhk.edu.cn)  \nChuanhui Yang  \nOceanBase, AntGroup Hangzhou, China [rizhao.ych@oceanbase.com](rizhao.ych@oceanbase.com)  \nFangyuan Zhang  \nThe Chinese University of Hong Kong Hong Kong, China [zzzzzfy@link.cuhk.edu.hk](zzzzzfy@link.cuhk.edu.hk)  \nYixiang Fang  \nThe Chinese University of Hong Kong, Shenzhen Shenzhen, China [fangyixiang@cuhk.edu.cn](fangyixiang@cuhk.edu.cn)  \narXiv :2607 .08539v 1 [ cs .DB] 9 Jul 2026  \nAbstract  \nLeveraging large language models (LLMs) to analyze complex documents — such as academic papers, technical manuals, and financial reports — has emerged as a mainstream and critical task in both research and industry. In practice, users must first filter relevant documents from large collections and then conduct in-depth analysis (e.g., question answering) over the selected subset, yet existing systems flatten documents into plain-text chunks, discarding the rich hierarchical structures (sections, tables, figures, equations) and degrading downstream performance. We present DocMaster, a hierarchical structure-aware document analysis system. DocMaster parses documents into hierarchical document trees preserving original layouts and constructs a structure-aware semantic index that enables accurate document filtering and in-depth analysis. We demonstrate DocMaster through an interactive web interface that enables users to upload document collections, construct treebased and multi-view semantic indices, filter relevant documents via natural-language conditions, and perform follow-up question answering over the filtered results. The source code, data, and demo are available at [https://doc-master.github.io/](https://doc-master.github.io/) .  \n1 Introduction  \nLeveraging large language models (LLMs) to analyze complex documents — such as academic papers, technical manuals, and financial reports — has emerged as a mainstream and critical task in both research and industry. These documents encode information within rich hierarchical structures —sections, subsections, tables, figures, and equations — that must be understood for effective analysis. In practice, users must first filter relevant documents from large collections and then conduct in-depth analysis (e.g., question answering) over the selected subset, as shown in Figure 1 . For example, a researcher surveying AI papers may need to find those that “propose a retrieval-augmented method and evaluate on open-domain QA benchmarks” before extracting detailed insights.  \nHowever, existing document analysis systems [4, 7, 9] flatten documents into plain-text chunks, discarding the rich hierarchical  \nFigure 1: The overall workflow of document analysis.  \nstructures and degrading downstream performance. This structureagnostic treatment introduces three key challenges. (C1) Hierarchy preservation. Flattening sections, subsections, and heterogeneous elements (text, tables, figures, equations) into flat chunks destroys structural relationships, causing systems to return contextpoor fragments that hinder accurate filtering and answering. (C2) Cross-section semantic indexing. Implicit semantic relationships span across sections—e.g., a method in Section 3 may be closely tied to an evaluation metric in Section 5 . Capturing such cross-section links requires indices that go beyond simple vector similarity. (C3) Cross-section evidence aggregation. A single filter condition may require evidence scattered across distant sections, yet existing systems retrieve chunks independently without aggregating cross-section evidence.  \nTo address the","cbCainNJwpkXZfze","https://ap.wps.com/l/cbCainNJwpkXZfze","pdf",1373559,2,1,4,"English","en",105,"# Introduction\n## System Overview\n### Document Parsing\n### Document Tree Construction\n## Semantic Index Construction\n## Filtering and Retrieval","[{\"question\":\"What problem does DocMaster address in document analysis with LLMs?\",\"answer\":\"It addresses the degradation caused by flattening documents into plain-text chunks, which discards hierarchical structure and weakens downstream filtering and question answering performance.\"},{\"question\":\"How does DocMaster preserve a document’s structure?\",\"answer\":\"It parses PDFs into hierarchical document trees that preserve original layouts and heterogeneous elements such as text blocks, tables, figures, and equations.\"},{\"question\":\"How does DocMaster filter relevant documents before answering questions?\",\"answer\":\"It uses a tri-modal retrieval strategy combining document-tree traversal, embedding-based semantic search, and hyper-edge matching, then performs retrieval-augmented generation over the filtered results for follow-up QA.\"}]",1784187202,10,{"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":85,"head_meta":87,"extra_data":89,"updated_unix":28},"docmaster-a-hierarchical-structure-aware-system-for-document-analysis","",{"@graph":36,"@context":84},[37,52,67],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":22},"https://docshare.wps.com/document/docmaster-a-hierarchical-structure-aware-system-for-document-analysis/83394/",{"url":51,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":24,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":41,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-23","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What problem does DocMaster address in document analysis with LLMs?","Question",{"text":74,"@type":75},"It addresses the degradation caused by flattening documents into plain-text chunks, which discards hierarchical structure and weakens downstream filtering and question answering performance.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does DocMaster preserve a document’s structure?",{"text":79,"@type":75},"It parses PDFs into hierarchical document trees that preserve original layouts and heterogeneous elements such as text blocks, tables, figures, and equations.",{"name":81,"@type":72,"acceptedAnswer":82},"How does DocMaster filter relevant documents before answering questions?",{"text":83,"@type":75},"It uses a tri-modal retrieval strategy combining document-tree traversal, embedding-based semantic search, and hyper-edge matching, then performs retrieval-augmented generation over the filtered results for follow-up QA.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,133],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":29,"doc_module":4,"doc_module_name":46,"category_name":131,"show_sort_weight":29,"slug":132},"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":46,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]