[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86263-en":3,"doc-seo-86263-105":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},86263,687197207919,"Theodora","https://ap-avatar.wpscdn.com/avatar/a000253d6f5f7c60be?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779446848396160552",8,"Research & Report","PaperRouter-Agent: A Content-Grounded LLM Agent for Personalized Hierarchical Paper Routing","Personalized hierarchical paper routing (PHPR) assigns incoming papers to folders in an individual’s evolving reference-manager hierarchy, where folder meanings come from existing members rather than names. PaperRouter-Agent is a training-free LLM agent that narrows candidates, retrieves folder-specific evidence, verifies fit by inspecting member papers, and uses similarity-gated feedback from prior user rejections. Experiments on real personal libraries improve Recall@1 from 0.39 to 0.61 and Recall@3 from 0.57 to 0.83, and on LaMP-2 boost accuracy from 44.5% to 51.5% (+9.0 macro-F1).","arXiv :2607 . 1 1564v 1 [ cs .CL] 13 Jul 2026  \nPaperRouter-Agent: A Content-Grounded LLM Agent for Personalized Hierarchical Paper  \nRouting  \nKeshen Zhou, Lintao Wang, Suqin Yuan, Zhuqiang Lu, Yu Luo, and  \nZhiyong Wang  \nUniversity of Sydney, Sydney, NSW 2006, Australia {kzho6770,lwan3720,zhlu6105,[yluo0465}@uni.sydney.edu.au](yluo0465}@uni.sydney.edu.au)[ ](yluo0465}@uni.sydney.edu.au)[suqinyuan.cs@gmail.com](suqinyuan.cs@gmail.com) , [zhiyong.wang@sydney.edu.au](zhiyong.wang@sydney.edu.au)  \nAbstract. Researchers organize the papers they collect into personal folder hierarchies in reference managers, and route each new paper into the folder where it belongs. This task differs from standard hierarchical text classification. A user’s folder hierarchy is not a fixed, shared taxonomy but a private and evolving folksonomy whose folder meanings may be topical, shorthand, venue-based, or process-oriented, and are often defined by the papers already stored inside them. We formalize this setting as personalized hierarchical paper routing (PHPR): assigning an incoming paper to folders in a user-specific hierarchy without peruser training. We propose PaperRouter-Agent, a training-free LLM agent that grounds routing decisions in folder members rather than folder names alone. The agent first narrows the candidate hierarchy, retrieves folder-specific evidence, verifies fit by inspecting member papers, and incorporates similarity-gated feedback from past user rejections. A formative study on real personal libraries shows that PaperRouter-Agent raises overall Recall@1 from 0.39 to 0.61 and Recall@3 from 0.57 to 0.83 , with the largest gains on organizational folders defined by metadata such as venue or year, where single-shot methods collapses (Recall@1 0.09 → 0 .50) . On the public LaMP-2 benchmark, the same approach improves accuracy from 44.5% to 51.5%(+9 .0 macro-F1) over a single-shot baseline, while remaining low-cost for practical use.  \nKeywords: Personalized text classification · Retrieval-augmented generation · LLM agents · Personal knowledge management  \n1 Introduction  \nOrganizing papers collected from the Web has become a routine burden of research practice. ArXiv’s monthly submissions have grown from 24k in October 2024 to a recent peak of 30k in March 2026 [1], and surveys of reading behavior find that researchers read on the order of 20 scholarly articles per month [29] . The papers worth keeping are added to personal reference managers such as Zotero and EndNote, where a library grows over years into a hierarchy of folders  \n2 K. Zhou et al.  \nFig. 1. Example: Comparison of a name-based baseline and the proposed contentgrounded PaperRouter-Agent. While the baseline misclassifies a multimodal paper based solely on its title, the proposed agent accurately routes it to appropriate topical and metadata-defined folders.  \ncontaining hundreds of papers. New papers keep arriving and the hierarchy itself keeps changing, so routing each paper into the right folder is not a one-time effort but a recurring task that requires recognizing which existing folder already holds similar work. Reference managers automate storage, search, and citation, but not this routing: which folder a paper belongs to is still determined by the user. Automating this step is a personalized classification problem: the labels are each researcher’s own folders, not a single schema shared across users.  \nThe difficulty of routing is that a folder’s meaning is defined by the papers a user has placed in it, whereas its name is often the only signal available for an incoming paper. In a personal library, folder names rarely follow a single organizing principle: some are topical (“Diffusion Models in Medical Imaging”), some are personal shorthand (“3D”, “MoE”), and some record a principle other than topic, such as a venue or a reading status (“CVPR”,“To Read”) . A zero-shot baseline that relies on names alone misclassifies CLIP [22] because of this gap, as Fig. ","cbCaih8j9DsNRHw0","https://ap.wps.com/l/cbCaih8j9DsNRHw0","pdf",3449667,3,1,24,"English","en",105,"# Introduction\n## Personalized hierarchical paper routing (PHPR)\n## PaperRouter-Agent overview\n## Routing challenges in personal folder hierarchies\n# Method\n## Candidate narrowing (Planner)\n## Evidence retrieval (Retriever)\n## Content-grounded verification (Inspector)\n## Similarity-gated feedback","[{\"question\":\"什么是个性化分层论文路由（PHPR）？\",\"answer\":\"PHPR是在用户个人且会不断变化的文件夹层级中，将新论文分配到对应文件夹。任务区别于固定标签体系，因为文件夹的含义主要由用户已存入的成员论文决定。\"},{\"question\":\"PaperRouter-Agent如何实现无需训练的路由决策？\",\"answer\":\"它先缩小候选层级集合，再按文件夹类型检索相应证据（主题类取内容证据，场馆/状态类取元数据证据），随后通过检查成员论文进行适配验证，并结合用户历史拒绝信息进行相似度门控反馈。\"},{\"question\":\"实验结果表明该方法相对基线提升在哪里？\",\"answer\":\"在真实个人文库研究中，Recall@1从0.39提升到0.61，Recall@3从0.57提升到0.83；在LaMP-2基准上，准确率从44.5%提升到51.5%（+9.0 macro-F1），同时保持较低的实际使用成本。\"}]",1784209903,60,{"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},"paperrouter-agent-a-content-grounded-llm-agent-for-personalized-hierarchical-paper-routing","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"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":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/paperrouter-agent-a-content-grounded-llm-agent-for-personalized-hierarchical-paper-routing/86263/",4,{"url":51,"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","2026-07-24","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},"什么是个性化分层论文路由（PHPR）？","Question",{"text":75,"@type":76},"PHPR是在用户个人且会不断变化的文件夹层级中，将新论文分配到对应文件夹。任务区别于固定标签体系，因为文件夹的含义主要由用户已存入的成员论文决定。","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"PaperRouter-Agent如何实现无需训练的路由决策？",{"text":80,"@type":76},"它先缩小候选层级集合，再按文件夹类型检索相应证据（主题类取内容证据，场馆/状态类取元数据证据），随后通过检查成员论文进行适配验证，并结合用户历史拒绝信息进行相似度门控反馈。",{"name":82,"@type":73,"acceptedAnswer":83},"实验结果表明该方法相对基线提升在哪里？",{"text":84,"@type":76},"在真实个人文库研究中，Recall@1从0.39提升到0.61，Recall@3从0.57提升到0.83；在LaMP-2基准上，准确率从44.5%提升到51.5%（+9.0 macro-F1），同时保持较低的实际使用成本。","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":25},{"code":4,"msg":5,"data":92},[93,97,101,105,109,114,119,122,127,130,134],{"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":52,"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":29,"slug":108},5,"Comic","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":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]