[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-427364-105":59,"doc-detail-427364-en":130},{"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":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","exploring-llm-based-generative-recommender-systems-corpora-customization-and-evaluation-insights","EXPLORING LLM-BASED GENERATIVE RECOMMENDER SYSTEMS: CORPORA, - CUSTOMIZATION, AND EVALUATION INSIGHTS","","Large Language Model-Driven Generative Recommender Systems (LLM-GRSs) are increasingly transforming healthcare, especially for question-answering and medical information delivery. This systematic review mapped corpora sources, customization strategies, and evaluation metrics across 61 studies from 2021–2024. Corpus sources included real-world clinical resources, literature materials, open-source datasets, and web-crawled data. Customization methods covered pre-training, prompt engineering, RAG, fine-tuning, in-context learning, and offline learning, while evaluation metrics spanned process, usability, and outcome domains. Key gaps include corpus fairness, biases by geography and socio-economic factors, and limited integration of verified clinical guidance.",{"@graph":69,"@context":122},[70,84,105],{"@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/exploring-llm-based-generative-recommender-systems-corpora-customization-and-evaluation-insights/427364/",{"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/exploring-llm-based-generative-recommender-systems-corpora-customization-and-evaluation-insights/427364.png","ImageObject",300,407,{"name":92,"@type":93},"Liam","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-16","2026-09-29",true,{"@type":102,"interactionType":103,"userInteractionCount":44},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What corpora sources were included in the reviewed LLM-GRS studies?","Question",{"text":112,"@type":113},"The review categorized corpus sources into real-world clinical resources, literature materials, open-source datasets, and web-crawled data.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"Which customization techniques were most commonly used?",{"text":117,"@type":113},"Reported strategies included pre-training, prompt engineering, retrieval-augmented generation (RAG), fine-tuning, in-context learning, and offline learning, with many studies combining multiple methods.",{"name":119,"@type":110,"acceptedAnswer":120},"How were evaluation metrics organized in the review?",{"text":121,"@type":113},"Evaluation metrics were grouped into process metrics, usability metrics, and outcome metrics, with outcome metrics further separated into model-based and expert-assessed results.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},427364,1790894908,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":44,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":14,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":143,"read_time":24},8796095461564,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","658 Innovation in Aging, 2025, Vol. 9, No. S2  \nAbstract citation ID: igaf122.2248  \nEXPLORING LLM-BASED GENERATIVE  \nRECOMMENDER SYSTEMS: CORPORA,  \nCUSTOMIZATION, AND EVALUATION INSIGHTS  \nShuqi Yang1, Mingrui Jing1, Shuai Wang2, Jiaqing Wang1, Weijie Xing1, Yan Hu1, and Zheng Zhu1, 1. Fudan University, Shanghai, Shanghai, China (People’s Republic), 2. Dali University, Dali, Shanghai, China (People’s Republic)  \nLarge Language Model-Driven Generative Recommender Systems (LLM-GRSs) are increasingly transforming healthcare, particularly in question-answering systems. This study systematically reviewed their corpora sources, customization techniques, and evaluation metrics. A search of PubMed/MEDLINE, Embase, Scopus, and Web of Science identified 61 studies (2021–2024) using LLM-GRSs for medical information delivery. Corpus sources were categorized into real-world clinical resources (n = 24), literature materials (n = 34), open-source datasets (n = 33), and web-crawled data (n = 11), with 44 studies  \nInnovation in Aging, 2025, Vol. 9, No. S2 659  \nintegrating multiple sources. Key model customization strategies included pre-training, prompt engineering, retrieval-augmented generation (RAG), fine-tuning, in-context learning, and offline learning. Fourteen studies used a single customization technique, while 41 studies combined these methods during model development. The evaluation metrics were classified into three main domains: 1) process metrics, 2) usability metrics, and 3) outcome metrics. The outcome metrics could also be divided into two categories: model-based outcomes and expert-assessed outcomes. The study identified critical gaps in corpus fairness, contributing to biases from geographic, cultural, and socio-economic factors. The reliance on unverified or unstructured data highlights the need for better integration of evidence-based clinical guidelines. Future research should focus on developing a tiered corpus architecture with vetted sources and dynamic weighting, while ensuring model transparency.  \nAdditionally, the lack of standardized evaluation frameworks for domain-specific models calls for comprehensive validation of LLM-GRSs in real-world healthcare settings.","cbCaikIPp8sQSraj","https://ap.wps.com/l/cbCaikIPp8sQSraj","pdf",55276,"English","# Systematic review scope and study identification\n## Corpus sources\n## Customization strategies\n## Evaluation metrics\n## Identified gaps and future directions","[{\"question\":\"What corpora sources were included in the reviewed LLM-GRS studies?\",\"answer\":\"The review categorized corpus sources into real-world clinical resources, literature materials, open-source datasets, and web-crawled data.\"},{\"question\":\"Which customization techniques were most commonly used?\",\"answer\":\"Reported strategies included pre-training, prompt engineering, retrieval-augmented generation (RAG), fine-tuning, in-context learning, and offline learning, with many studies combining multiple methods.\"},{\"question\":\"How were evaluation metrics organized in the review?\",\"answer\":\"Evaluation metrics were grouped into process metrics, usability metrics, and outcome metrics, with outcome metrics further separated into model-based and expert-assessed results.\"}]","EXPLORING LLM-BASED GENERATIVE RECOMMENDER SYSTEMS: CORPORA, - CUSTOMIZATION, AND EVALUATION INSIGHTS | PDF",1790643657]