[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86071-en":3,"doc-seo-86071-105":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},86071,687197207057,"Sage","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","ZoRRO: A Zero-Weight Personalized Recommender System for Scalable News Recommendation","ZoRRO (Zero-Weight Personalized Recommender System) introduces a zero-weight, training-free framework for personalized news recommendation aimed at scalable real-world deployment. It improves offline ranking results over strong neural baselines and achieves online click-through rate performance nearly matching a state-of-the-art deep model, while delivering over 600× faster serving. Experiments expose offline–online gaps and show similar CTR can yield substantially different recommendation distributions. The work emphasizes evaluating recommender systems beyond accuracy and provides open-source code.","ZoRRO: A Zero-Weight Personalized Recommender System for  \nScalable News Recommendation  \nJohannes Kruse∗ [jkru@dtu.dk](jkru@dtu.dk)  \nTechnical University of Denmark Kongens Lyngby, Denmark  \nRyotaro Shimizu  \n[ryotaro.shimizu@zozo.com](ryotaro.shimizu@zozo.com)[ ](ryotaro.shimizu@zozo.com)ZOZO Research Tokyo, Japan  \nKasper Lindskow  \n[kasper.lindskow@jppol.dk](kasper.lindskow@jppol.dk)[ ](kasper.lindskow@jppol.dk)[JP/Politikens Media Group](JP/Politikens Media Group)[ ](JP/Politikens Media Group)Copenhagen, Denmark  \nJon Tofteskov  \n[jon.tofteskov@jppol.dk](jon.tofteskov@jppol.dk)[ ](jon.tofteskov@jppol.dk)[JP/Politikens Media Group](JP/Politikens Media Group)[ ](JP/Politikens Media Group)Copenhagen, Denmark  \nMichael Riis Andersen  \n[miri@dtu.dk](miri@dtu.dk)  \nTechnical University of Denmark Kongens Lyngby, Denmark  \nJulian McAuley  \n[jmcauley@eng.ucsd.edu](jmcauley@eng.ucsd.edu)[ ](jmcauley@eng.ucsd.edu)University of California San Diego La Jolla, California, USA  \nJes Frellsen† [jefr@dtu.dk](jefr@dtu.dk)  \nTechnical University of Denmark Kongens Lyngby, Denmark  \narXiv :2607 . 109 10v 1 [ cs .IR] 12 Jul 2026  \nAbstract  \nIn this paper, we present ZoRRO (Zero-Weight Personalized Recommender System), a zero-weight and training-free framework for personalized news recommendation designed for scalable realworld deployment. We show that ZoRRO outperforms strong neural baselines in offline ranking evaluations and delivers click-through rate performance in online A/B testing that is nearly on par with a state-of-the-art deep learning model, while operating more than 600× faster. Our experiments reveal gaps between offline and online performance, and show that models with similar click-through rate (CTR) outcomes can produce markedly different recommendation distributions, influencing the overall news flow. These findings position ZoRRO as a practical and efficient solution for large-scale news recommendation and highlight the importance of evaluating recommender systems using metrics beyond accuracy alone. Our code is available at [https://github.com/johanneskruse/zorro](https://github.com/johanneskruse/zorro).  \nCCS Concepts  \n• Information systems → Recommender systems.  \nKeywords  \nrecommender systems, news recommendation, scalability, online A/B testing  \n1 Introduction  \nUnlike other domains, news platforms face unique challenges due to the constant influx of new articles and their short lifespan. This dynamic environment demands recommendation systems that quickly adapt to fresh content and evolving user interests, making the coldstart problem especially severe for both items and users [5, 17– 19, 37] . Despite advances in deep learning-based recommender  \n∗ Also with JP/Politikens Media Group.  \n†Also with Pioneer Centre for Artificial Intelligence.  \nThis work is licensed under a Creative Commons Attribution 4 .0 International License.  \nsystems (DLRS), these methods often require substantial computational resources, specialized expertise, and continuous maintenance [3, 4, 10, 13, 30] . In practice, simpler approaches often prevail, as complex models are harder to implement and sustain, limiting adoption. Moreover, while DLRS offer great flexibility, their benefits and reported gains are not always justified [7, 21, 22, 27, 29] . Consequently, there is growing interest in practical, scalable approaches that balance effectiveness with operational efficiency [18]. News recommenders must efficiently handle rapid content turnover, address cold-start issues, and deliver timely, personalized recommendations without heavy computational demands [36, 37] .  \nTo meet these challenges, we propose ZoRRO (Zero-Weight Personalized Recommender System), a simple yet effective framework optimized for the dynamic nature of news recommendation. ZoRRO combines article recency with representation-based similarity and can incorporate article representations generated through different methods, including one-hot category encodings and pre-tr","cbCaidc3NFyu7xCp","https://ap.wps.com/l/cbCaidc3NFyu7xCp","pdf",653393,3,1,6,"English","en",105,"# Abstract\n# 1 Introduction\n# 2 Related work","[{\"question\":\"What is ZoRRO and what makes it different from typical recommender systems?\",\"answer\":\"ZoRRO is a zero-weight, training-free framework that combines article recency with representation-based similarity, avoiding costly retraining and heavy computation for serving.\"},{\"question\":\"How does ZoRRO perform in offline and online evaluation?\",\"answer\":\"ZoRRO outperforms strong neural baselines in offline ranking evaluations and delivers click-through rate in online A/B testing nearly on par with a state-of-the-art deep learning model, while running more than 600× faster at inference.\"},{\"question\":\"Why do the results highlight differences between offline metrics and online user impact?\",\"answer\":\"The study shows gaps between offline and online performance: systems with similar CTR outcomes can still produce notably different recommendation distributions, affecting the overall news flow.\"}]",1784208313,15,{"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},"zorro-a-zero-weight-personalized-recommender-system-for-scalable-news-recommendation","",{"@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/zorro-a-zero-weight-personalized-recommender-system-for-scalable-news-recommendation/86071/",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","2026-07-26","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 is ZoRRO and what makes it different from typical recommender systems?","Question",{"text":75,"@type":76},"ZoRRO is a zero-weight, training-free framework that combines article recency with representation-based similarity, avoiding costly retraining and heavy computation for serving.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does ZoRRO perform in offline and online evaluation?",{"text":80,"@type":76},"ZoRRO outperforms strong neural baselines in offline ranking evaluations and delivers click-through rate in online A/B testing nearly on par with a state-of-the-art deep learning model, while running more than 600× faster at inference.",{"name":82,"@type":73,"acceptedAnswer":83},"Why do the results highlight differences between offline metrics and online user impact?",{"text":84,"@type":76},"The study shows gaps between offline and online performance: systems with similar CTR outcomes can still produce notably different recommendation distributions, affecting the overall news flow.","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},[93,97,101,105,110,114,119,122,127,130,134],{"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":108,"slug":109},5,"Comic",60,"comic",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},"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"]