[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85395-en":3,"doc-seo-85395-105":29,"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":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85395,1099513958607,"Jiven","https://ap-avatar.wpscdn.com/avatar/100002390cf8733938c?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778829742770036399",8,"Research & Report","HiQA: A Hierarchical Contextual Augmentation RAG for Multi-Document QA","Retrieval-Augmented Generation (RAG) improves document-based question answering by using external documents during generation, but retrieval accuracy drops when many documents share similar semantics and structure. HiQA presents a hierarchical contextual augmentation framework for multi-document QA, enriching text chunks with cascading document metadata such as titles and section paths. It uses a multi-route retriever combining semantic, lexical, and keyword/entity signals. MasQA is introduced to benchmark realistic similar-document settings. Experiments show stronger gains on highly similar, structured domain collections.","arXiv :2402 .0 1767v 3 [ cs .CL] 13 Jul 2026  \nHiQA: A Hierarchical Contextual Augmentation RAG for  \nMulti-Document QA  \nXinyue Chen 1 , Pengyu Gao2 , Jiangjiang Song 1 , Xinjian Chen3 , Xiaoyang Tan 1*  \n1 Nanjing University of Aeronautics and Astronautics.  \n2 Southeast University, Nanjing, Jiangsu, China.  \n3 Hello World(Shanghai) Technology Co. , Ltd. , Shanghai, China.  \n*Corresponding author(s). E-mail(s): [x.tan@nuaa.edu.cn](x.tan@nuaa.edu.cn) ;  \nAbstract  \nRetrieval-Augmented Generation (RAG) significantly improves document-based question answering by integrating external documents during generation. However, retrieval accuracy can degrade when the knowledge base contains many semantically and structurally similar documents. We introduce HiQA, a practical hierarchical contextual augmentation framework for multi-document question answering (MDQA) . HiQA enriches text chunks with cascading document metadata, such as document titles and section paths, so that retrieval can use both local content and document structure. The framework also uses a multi-route retriever that combines semantic, lexical, and keyword/entity signals. We further introduce MasQA, a benchmark designed to evaluate MDQA systems in realistic similar-document settings.  \nExperiments show that HiQA improves retrieval and answer quality on MasQA and remains competitive on public MDQA benchmarks, while its benefits are strongest for structured, domain-specific, highly similar document collections.  \nKeywords: Retrieval-Augmented Generation,Multi-Document QA,Query Answering,NLP,Text  \nEnhancement,Fusion Retrieval,Large Language Models  \n1 Introduction  \nLarge Language Models (LLMs) have gained widespread popularity and accessibility, resulting in impressive applications across various domains [1–4] . One such domain is document-based question-answering (QA) [5– 7], driven by the significant demand for document reading among people or question-answering system in open-domain. Retrieval-Augmented Generation (RAG) is a promising solution to these applications [8] .  \nStandard RAG-based document QA systems represent documents as unstructured text chunks, then retrieve a few chunks to answer questions. This approach encounters limitations as document sizes increase, especially when dealing with documents that have similar and complex content or structures. Scaling laws commonly recommend that more data lead to better performance [2] . Nevertheless, as the number of documents increases, the accuracy of responses continuously declines. We identify this issue as ”RAG degradation in indistinguishable multi-document.” We illustrate this in Figure 1. As the number of documents increases, the signal-to-noise ratio decreases, making it challenging for existing methods to differentiate relevant information from noise. Although this issue has not received sufficient attention in existing literature, such scenarios are prevalent in real-world applications.  \nThis performance degradation primarily results from three key limitations of current retrieval approaches. First, conventional retrieval systems heavily rely on similarity-based metrics that fail to adequately capture the true contextual relevance of the retrieved information. Second, in contrast to traditional information retrieval systems—which frequently span diverse, general-purpose datasets and rely on human judgment for refinement—RAG systems often operate on domain-specific document collections curated by experts [10] . These specialized collections usually have high internal cohesion and low diversity, leading to substantial overlap and semantic similarity among documents. Consequently, retrieval systems struggle to differentiate contextually relevant information from semantically similar but irrelevant content [11, 12] . And third, human  \nFig. 1: Experimental validation of performance degradation in multi-document QA scenario. Testing with 88 documents, each containing one of 88 questions. Using a vanilla R","cbCaifkqiNWQ51ry","https://ap.wps.com/l/cbCaifkqiNWQ51ry","pdf",7906589,1,33,"English","en",105,"# Introduction\n## RAG degradation in indistinguishable multi-document\n## Key limitations of current retrieval\n## Metadata-enhanced retrieval with HiQA\n## Benchmarking with MasQA","[{\"question\":\"What problem does HiQA address in multi-document question answering?\",\"answer\":\"HiQA targets performance degradation when the knowledge base contains many semantically and structurally similar documents, making it harder to retrieve contextually relevant information.\"},{\"question\":\"How does HiQA improve retrieval accuracy?\",\"answer\":\"HiQA enriches chunks with hierarchical document metadata, such as document titles and section paths, and uses a multi-route retriever that fuses semantic, lexical, and keyword/entity signals.\"},{\"question\":\"What is MasQA and why is it introduced?\",\"answer\":\"MasQA is a benchmark designed to evaluate multi-document QA systems under realistic similar-document settings, where retrieval is especially challenging due to high overlap among documents.\"}]",1784203106,83,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"hiqa-a-hierarchical-contextual-augmentation-rag-for-multi-document-qa","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/hiqa-a-hierarchical-contextual-augmentation-rag-for-multi-document-qa/85395/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","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 HiQA address in multi-document question answering?","Question",{"text":75,"@type":76},"HiQA targets performance degradation when the knowledge base contains many semantically and structurally similar documents, making it harder to retrieve contextually relevant information.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does HiQA improve retrieval accuracy?",{"text":80,"@type":76},"HiQA enriches chunks with hierarchical document metadata, such as document titles and section paths, and uses a multi-route retriever that fuses semantic, lexical, and keyword/entity signals.",{"name":82,"@type":73,"acceptedAnswer":83},"What is MasQA and why is it introduced?",{"text":84,"@type":76},"MasQA is a benchmark designed to evaluate multi-document QA systems under realistic similar-document settings, where retrieval is especially challenging due to high overlap among documents.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]