[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84925-en":3,"doc-seo-84925-105":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},84925,1099514068035,"Ezra","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Hypothesis-Driven Model Expansion under Uncertainty for Open-World Robot Planning","The paper studies open-world robot planning for service robots operating in unknown household environments with incomplete knowledge of objects and actions. It addresses limitations of closed-world, predefined knowledge bases that break under unexpected tasks and situations. The proposed open-world framework automatically generates, verifies, and updates uncertainty-aware hypotheses about abstract world models. Hypotheses are produced using foundation models and validated through automated planning that jointly executes actions for verification and task completion. Iterative execution with refinement leverages verification feedback to expand knowledge in simulation and real-world experiments.","Hypothesis-driven Model Expansion under Uncertainty  \nfor Open-World Robot Planning  \nAnxing Xiao 1 , Hanbo Zhang 1,2 , Tianrun Hu 1,2 , David Hsu 1,2  \n1 School of Computing, National University of Singapore  \n2 Smart Systems Institute, National University of Singapore  \narXiv :2607 .0650 1v 1 [ cs .RO] 7 Jul 2026  \nAbstract—We consider an open-world planning setting in which service robots operate in unknown environments withincomplete knowledge of objects and actions. Traditional closedworld approaches with pre-programmed knowledge bases fail when robots encounter unexpected situations and tasks, posing a fundamental challenge for autonomous knowledge expansion inhuman environments. In this work, we propose an open-world planning framework that enables robots to automatically generate, verify, and update hypotheses about their abstract world models. Our key insight is to explicitly maintain uncertaintyaware knowledge expansion and integrate hypothesis verification into goal-reaching planning. Our framework, Hypothesisdriven Uncertainty-aware Model Expansion (HUME), leverages foundation models to generate initial hypotheses over states and transitions, and applies automated planning to produce action sequences that jointly address hypothesis verification and task execution. Through iterative execution and refinement, the robot expands its knowledge by incorporating verification feedback from the foundation models when hypotheses prove incorrect. Extensive experiments in simulated and real-world environments demonstrate that our framework enables autonomous knowledge expansion and effective operation in open-world settings. These results indicate that integrating uncertainty-aware model expansion from robot foundation models with planning advances the practical deployment of household service robots. Project website: [open-world-planning.github.io](open-world-planning.github.io)  \nI. INTRODUCTION  \nHome service robots operate under substantial uncertainty in real-world household environments, where objects may be hidden, their attributes unknown, and the effects of actions underspecified. Consider a daily task such as “serve a heated chicken burger”, as shown in Fig. 1. To accomplish this task, a robot must infer the burger’s location, determine whether it is made of chicken, and, in some cases, reason about appliance-specific action effects, such as how much the heating time increases when pressing the ‘+’ button on a microwave. These challenges make household environments inherently open-world. Classical automated planning relies on predefined state representations and transition models [5, 20, 21] . While effective in well-specified domains, this reliance on manually designed knowledge and closed-world assumptions limits generalization to open-world settings with underspecified knowledge. In contrast, Large Language Models (LLMs) exhibit remarkable reasoning and generalization to predict plausible plans without explicitly modeling states or transition dynamics [6, 18, 28, 29, 38] . Yet, in open-world settings with partial information, these properties can blur the boundary between reasoning grounded in known facts and reasoning based on hallucinated knowledge. This motivates a central question of  \n“Heat up a chicken burger”  \nHypotheses are uncertain: H1, H2, H3,    \nH1: may be inside the fridge  \nH3:  may add 10 seconds of heating time  \nH2: may be made of chicken  \nFig. 1. Illustration of a service robot operating in an open-world scenario. To fulfill tasks such as “heat up a chicken burger”, the robot must expand its abstract world model and account for uncertainty during planning.  \nthis work: how can we combine structured reasoning with the unstructured open-world knowledge of foundation models?  \nTo bridge this gap, recent research has explored using foundation models to construct or fix symbolic representations of the environment [1, 7, 17, 39, 42, 70–72] . These approaches allow models to generate planning domains","cbCaijMmYly3A08g","https://ap.wps.com/l/cbCaijMmYly3A08g","pdf",13450627,1,38,"English","en",105,"# Introduction\n## Open-world uncertainty in home service robotics\n## Motivation: uncertainty and hallucinations in foundation-model reasoning\n## Core question and proposed approach (HUME)\n## Bayes-adaptive perspective and uncertainty-aware hypothesis modeling","[{\"question\":\"What problem does the paper address in open-world robot planning?\",\"answer\":\"It addresses planning in unknown environments where robots have incomplete knowledge of objects and action effects, causing traditional closed-world planners to fail when encountering unexpected situations.\"},{\"question\":\"How does HUME handle uncertainty during model expansion?\",\"answer\":\"HUME treats hypotheses generated by foundation models as uncertain latent variables, explicitly tracking uncertainty and integrating hypothesis verification into goal-reaching planning rather than assuming generated models are correct.\"},{\"question\":\"What is the role of hypothesis verification in action planning?\",\"answer\":\"Automated planning produces action sequences that jointly support task execution and hypothesis verification, and the robot refines its model by incorporating verification feedback when hypotheses prove incorrect.\"}]",1784199366,96,{"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},"hypothesis-driven-model-expansion-under-uncertainty-for-open-world-robot-planning","",{"@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/hypothesis-driven-model-expansion-under-uncertainty-for-open-world-robot-planning/84925/",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 the paper address in open-world robot planning?","Question",{"text":75,"@type":76},"It addresses planning in unknown environments where robots have incomplete knowledge of objects and action effects, causing traditional closed-world planners to fail when encountering unexpected situations.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does HUME handle uncertainty during model expansion?",{"text":80,"@type":76},"HUME treats hypotheses generated by foundation models as uncertain latent variables, explicitly tracking uncertainty and integrating hypothesis verification into goal-reaching planning rather than assuming generated models are correct.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the role of hypothesis verification in action planning?",{"text":84,"@type":76},"Automated planning produces action sequences that jointly support task execution and hypothesis verification, and the robot refines its model by incorporating verification feedback when hypotheses prove incorrect.","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"]