[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85523-en":3,"doc-seo-85523-105":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},85523,34359740700684,"Finn","https://ap-avatar.wpscdn.com/avatar/1f400023980c374ae676?_k=1777273430885731487",8,"Research & Report","Driving Like Yourself: A Benchmark for Closed-Loop Personalized End-to-End Autonomous Driving","Human driving behavior varies by cognitive patterns, habits, and personality, yet most end-to-end autonomous driving systems learn a single average style and ignore individual differences. Person2Drive introduces a closed-loop personalized E2E-AD platform and benchmark, with scalable human-style trajectory data collection, style-vector evaluation using Maximum Mean Discrepancy and KL divergence, and a style reward model enabling plug-and-play personalization by fine-tuning only the trajectory prediction head to maintain safety while preserving performance and success rate in challenging scenarios.","arXiv :2602 . 18757v3 [ cs .CV] 13 Jul 2026  \nDriving Like Yourself: A Benchmark for Closed-Loop Personalized End-to-End Autonomous Driving  \nXiaoru Dong 1 ,2†, Ruiqin Li2†, Xiao Han2†, Zhenxuan Wu2, Jiamin Wang2, Jian Chen 1, Qi Jiang2, Siu Ming Yiu 1∗ , Xinge Zhu3, and  \nYuexin Ma2∗  \n1 The University of Hong Kong  \n2 ShanghaiTech University  \n3 The Chinese University of Hong Kong  \nxrdong@cs.hku.hk, [mayuexin@shanghaitech.edu.cn](mayuexin@shanghaitech.edu.cn)  \nAbstract. Human driving behavior is inherently diverse, yet most endto-end autonomous driving (E2E-AD) systems learn a single average driving style, neglecting individual differences. Achieving personalized E2E-AD faces challenges across three levels: limited real-world datasets with individual-level annotations, a lack of quantitative metrics for evaluating personal driving styles, and the absence of algorithms that can learn stylized representations from users’ trajectories. To address these gaps, we propose Person2Drive, a comprehensive personalized E2E-AD platform and benchmark. It includes an open-source, flexible data collection system that simulates realistic scenarios to generate scalable, diverse personalized driving datasets; style vector–based evaluation metrics with Maximum Mean Discrepancy and KL divergence to comprehensively quantify individual driving behaviors; and a personalized E2E-AD framework with a style reward model that efficiently adapts E2E models for safe and individualized driving. Crucially, our framework enables plug-and-play personalization by fine-tuning only the trajectory prediction head, preserving the pretrained base model and ensuring safety. Extensive experiments demonstrate that Person2Drive enables fine-grained analysis and effective personalization, while preserving driving performance and success rate even in challenging scenarios.  \n1 Introduction  \nTraditional autonomous driving systems rely on modular pipelines that separate perception, prediction, and control, often leading to suboptimal coordination, whereas recent end-to-end (E2E-AD) approaches use unified deep learning models to map sensor inputs directly to control outputs, achieving notable improvements in behavioral consistency and safety [1,2,7] . However, most E2E-AD  \n∗ Corresponding authors.† Equal contribution.  \n2 Dong et al.  \napproaches employ a single model to fit all drivers, effectively learning an average driving style while neglecting individual differences. In reality, human driving behavior is inherently diverse—shaped by cognitive patterns, habits, and personality traits [24] . Overlooking such individuality limits our understanding of human decision-making and impedes the development of human-centered, trustworthy driving systems [13] . Personalized E2E-AD, by contrast, aims to adapt driving behavior to individual preferences and traits, thereby improving comfort, trust, and user acceptance [8, 11, 12] .  \nTo address personalization in autonomous driving, existing methods can be roughly divided into three categories. Early approaches introduce rule-based adjustments or trajectory-level tuning on modular pipelines [10], which are inherently incompatible with end-to-end learning. With the rise of E2E-AD, some studies condition models on discrete driving-style labels (e.g. , aggressive, normal, conservative) [9] . However, human driving behaviors vary continuously across scenarios, making such coarse categorization insufficient for capturing individual dynamics. More recently, language-conditioned control has been explored for interpretable human–vehicle interaction [4, 27, 32], yet language signals alone cannot represent fine-grained, continuous control characteristics. This calls for anew personalized E2E-AD paradigm capable of adapting driving styles directly from users’ trajectory data.  \nAgainst this background, significant challenges remain in achieving truly individual-level personalized E2E-AD. These challenges arise throughout the entire proces","cbCaipFbJZ8UsVjv","https://ap.wps.com/l/cbCaipFbJZ8UsVjv","pdf",2473237,4,1,18,"English","en",105,"# Introduction\n## Challenges in Personalized E2E-AD\n## Data Level Limitations\n## Evaluation Level Limitations\n## Algorithm Level Limitations\n## Person2Drive Contributions\n# Person2Drive Framework Overview\n## Personal Driving Data Collection\n## Diverse Human Data\n## Diverse Routes","[{\"question\":\"Why do current end-to-end autonomous driving systems underperform for personalization?\",\"answer\":\"Most systems fit a single model to all drivers, effectively learning an average driving style and neglecting individual differences in driving behavior.\"},{\"question\":\"What are the three main challenges Person2Drive targets?\",\"answer\":\"Challenges span the data level (limited datasets and missing individual-level annotations), the evaluation level (lack of standardized quantitative metrics), and the algorithm level (difficulty learning stylized representations from trajectories with a unified end-to-end framework).\"},{\"question\":\"How does Person2Drive enable plug-and-play personalization while preserving safety?\",\"answer\":\"The approach fine-tunes only the trajectory prediction head, keeping the pretrained base model intact, and uses a style reward model to adapt driving behavior for safe, individualized driving.\"}]",1784204158,45,{"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},"driving-like-yourself-a-benchmark-for-closed-loop-personalized-end-to-end-autonomous-driving","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"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":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/driving-like-yourself-a-benchmark-for-closed-loop-personalized-end-to-end-autonomous-driving/85523/",{"url":52,"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},"Why do current end-to-end autonomous driving systems underperform for personalization?","Question",{"text":75,"@type":76},"Most systems fit a single model to all drivers, effectively learning an average driving style and neglecting individual differences in driving behavior.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What are the three main challenges Person2Drive targets?",{"text":80,"@type":76},"Challenges span the data level (limited datasets and missing individual-level annotations), the evaluation level (lack of standardized quantitative metrics), and the algorithm level (difficulty learning stylized representations from trajectories with a unified end-to-end framework).",{"name":82,"@type":73,"acceptedAnswer":83},"How does Person2Drive enable plug-and-play personalization while preserving safety?",{"text":84,"@type":76},"The approach fine-tunes only the trajectory prediction head, keeping the pretrained base model intact, and uses a style reward model to adapt driving behavior for safe, individualized driving.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"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":20,"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":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]