[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85592-en":3,"doc-seo-85592-105":30,"detail-sidebar-cat-0-en-105":92},{"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},85592,1649267921044,"Ava Thompson","https://us-avatar.wpscdn.com/avatar/1800007509477c92dfb?_k=1782875107921204101",8,"Research & Report","EmbodiSkill Skill-Aware Reflection for Self-Evolving Embodied Agents","Embodied agents rely on skills to guide object search, action execution, and state transitions across diverse environments. Because environment layouts, object states, and feasible action sequences vary, skills must self-evolve from task trajectories. Existing self-evolution often performs coarse, trajectory-to-skill conversions, causing failed executions to be misread as skill defects. EmbodiSkill proposes a training-free framework using skill-aware reflection and targeted revision to update skill bodies with skill-changing evidence while preserving guidance using execution-lapse evidence. Experiments on ALFWorld and EmbodiedBench show improved success.","arXiv :2605 . 10332v2 [ cs .AI] 11 Jul 2026  \n2026-7-14  \nEmbodiSkill: Skill-Aware Reflection for Self-Evolving Embodied Agents  \nRuofei Ju1∗† Xinrui Wang2∗† Xin Ding3‡ Yifan Yang4 Hao Wu1 Shiqi Jiang4 Qianxi Zhang4 Hao Wen5 Xiangyu Li5 Weijun Wang5 Kun Li5 Yunxin Liu5 Haipeng Dai1 Wei Wang1 Ting Cao5‡  \n1 Nanjing University 2 Huazhong University of Science and Technology 3 University of Science and Technology of China  \n4 Microsoft Research 5 Institute for AI Industry Research (AIR), Tsinghua University  \nEmbodied agents can benefit from skills that guide object search, action execution, and state changes across diverse environments. Since embodied environments vary across layouts, object states, and other execution factors, these skills must self-evolve from trajectories generated during task execution. However, existing skill self-evolution methods are mainly developed in digital environments and often convert trajectories into coarse skill updates. Directly applying this paradigm to embodied settings is problematic, because a failed task execution may reflect not only incorrect skill content, but also an execution lapse in which the agent fails to follow valid guidance. We propose EmbodiSkill, a training-free framework for embodied skill self-evolution through skill-aware reflection and targeted revision. EmbodiSkill interprets each trajectory with respect to the current skill, uses skill-changing evidence to update the skill body, and uses executionlapse evidence to preserve and emphasize valid guidance. Experiments on ALFWorld and EmbodiedBench show that EmbodiSkill consistently improves embodied task success. On ALFWorld, EmbodiSkill enables a frozen Qwen3.5-27B executor to reach 93.28% task success, outperforming GPT-5.2 used as a direct agent without skills by 31.58%. These results show that skill-aware self-evolution helps embodied agents accumulate reusable procedural knowledge from their own trajectories. The code and data are available at [https://github.com/air-embodied-brain/EmbodiSkill](https://github.com/air-embodied-brain/EmbodiSkill).  \n1. Introduction  \nEmbodied agents are expected to complete household-like tasks in physical or physically grounded 3D environments, where they must observe the scene, navigate through space, interact with objects, satisfy action preconditions, and handle failed actions [1, 2, 3] . In such tasks, the trajectories generated by different task executions often exhibit recurring procedural structures: an agent may need to locate an object before manipulating it, check whether a container is open before placing it, or adjust its viewpoint when the target is not visible. Following prior work that treats skills, action primitives, or programmatic policies as reusable procedural units for grounding high-level instructions into executable behavior [4, 5, 6, 7], we define an embodied skill as a persistent and revisable procedural specification that guides an embodied agent across task executions. In embodied tasks, a skill specifies reusable guidance such as prerequisites, subgoal ordering, object affordances, visual-search strategies, action preconditions, and recovery strategies. Since embodied environments vary in layouts, object states, visibility conditions, and feasible action sequences, an initial skill cannot cover all situations the agent may encounter. Therefore, the central challenge is not only how an embodied agent uses a skill, but how the skill can self-evolve from trajectories into a more complete, accurate, and executable form.  \nRecent skill self-evolution methods show that agents can improve by extracting, revising, and reusing skills from trajectories [7, 8, 9, 10, 11, 12] . However, many existing methods update skills at a coarse  \n*  \nRuofei Ju and Xinrui Wang contributed equally to this work.  \n†This work was done while Ruofei Ju and Xinrui Wang were interns at the Institute for AI Industry Research (AIR), Tsinghua University.  \n‡Corresponding authors: Xin Ding ([","cbCaiqwW0UAVNQg8","https://ap.wps.com/l/cbCaiqwW0UAVNQg8","pdf",685342,6,1,15,"English","en",105,"# Introduction\n## Embodied skill concept and self-evolution challenge\n## Limitations of existing coarse update methods\n## Proposed framework: skill-aware reflection and targeted revision","[{\"question\":\"What problem does EmbodiSkill address in embodied skill self-evolution?\",\"answer\":\"It addresses how to make skills self-evolve from trajectories when failures may stem from an execution lapse rather than incorrect skill content, which undermines coarse skill update methods.\"},{\"question\":\"How does EmbodiSkill revise skills differently from prior approaches?\",\"answer\":\"It performs skill-aware reflection on trajectories: it uses skill-changing evidence to update the skill body and uses execution-lapse evidence to preserve and emphasize valid guidance.\"},{\"question\":\"What experimental results demonstrate the effectiveness of EmbodiSkill?\",\"answer\":\"On ALFWorld, EmbodiSkill enables a frozen Qwen3.5-27B executor to achieve 93.28% task success, outperforming a skills-free GPT-5.2 direct agent by 31.58%. Experiments on EmbodiedBench also show consistent improvements.\"}]",1784204788,38,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"embodiskill-skill-aware-reflection-for-self-evolving-embodied-agents","",{"@graph":36,"@context":86},[37,54,69],{"@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":53},"https://docshare.wps.com/document/embodiskill-skill-aware-reflection-for-self-evolving-embodied-agents/85592/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does EmbodiSkill address in embodied skill self-evolution?","Question",{"text":76,"@type":77},"It addresses how to make skills self-evolve from trajectories when failures may stem from an execution lapse rather than incorrect skill content, which undermines coarse skill update methods.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does EmbodiSkill revise skills differently from prior approaches?",{"text":81,"@type":77},"It performs skill-aware reflection on trajectories: it uses skill-changing evidence to update the skill body and uses execution-lapse evidence to preserve and emphasize valid guidance.",{"name":83,"@type":74,"acceptedAnswer":84},"What experimental results demonstrate the effectiveness of EmbodiSkill?",{"text":85,"@type":77},"On ALFWorld, EmbodiSkill enables a frozen Qwen3.5-27B executor to achieve 93.28% task success, outperforming a skills-free GPT-5.2 direct agent by 31.58%. Experiments on EmbodiedBench also show consistent improvements.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},"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":107,"slug":138},19,"General","general"]