{"id":8229,"date":"2026-02-17T05:27:00","date_gmt":"2026-02-16T20:27:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=8229"},"modified":"2026-02-15T14:29:30","modified_gmt":"2026-02-15T05:29:30","slug":"ui-mem-self-evolving-experience-memory-for-online-reinforcement-learning-in-mobile-gui-agents","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=8229","title":{"rendered":"UI-Mem: Self-Evolving Experience Memory for Online Reinforcement Learning in Mobile GUI Agents"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>UI-Mem: Self-Evolving Experience Memory for Online Reinforcement Learning in Mobile GUI Agents\u00a0<\/strong>[50.1]<br>\u30aa\u30f3\u30e9\u30a4\u30f3\u5f37\u5316\u5b66\u7fd2(RL)\u306f\u3001\u76f4\u63a5\u7684\u306a\u74b0\u5883\u76f8\u4e92\u4f5c\u7528\u3092\u901a\u3058\u3066GUI\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092\u5f37\u5316\u3059\u308b\u305f\u3081\u306e\u6709\u671b\u306a\u30d1\u30e9\u30c0\u30a4\u30e0\u3092\u63d0\u4f9b\u3059\u308b\u3002 \u968e\u5c64\u7684\u30a8\u30af\u30b9\u30da\u30ea\u30a8\u30f3\u30b9\u30e1\u30e2\u30ea\u306b\u3088\u308bGUI\u30aa\u30f3\u30e9\u30a4\u30f3RL\u3092\u5f37\u5316\u3059\u308b\u65b0\u3057\u3044\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3067\u3042\u308bUI-Mem\u3092\u63d0\u6848\u3059\u308b\u3002 UI-Mem\u306f\u5f93\u6765\u306eRL\u30d9\u30fc\u30b9\u30e9\u30a4\u30f3\u3084\u9759\u7684\u518d\u5229\u7528\u6226\u7565\u3088\u308a\u3082\u5927\u5e45\u306b\u512a\u308c\u3066\u3044\u307e\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2602.05832v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2602.05832v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 05 Feb 2026 16:21:43 GMT)<\/li>\n\n\n\n<li>\u300cconstructs a hierarchical, self-evolving memory that decom- poses raw experiences into reusable workflows, subtask skills, and failure patterns. We utilized this memory through a stratified group sampling mechanism tailored for GRPO, which balances memory-guided exploitation with necessary exploration to facilitate effective advantage estimation.\u300d\u3068GUI\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u305f\u3081\u306e\u30e1\u30e2\u30ea\u6a5f\u80fd\u63d0\u6848\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/ui-mem.github.io\/\">UI-Mem: Self-Evolving Experience Memory for Online Reinforcement Learning in Mobile GUI Agents<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[181,244,356],"class_list":["post-8229","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-gui-agent","tag-memory","tag-self-x"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/8229","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=8229"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/8229\/revisions"}],"predecessor-version":[{"id":8230,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/8229\/revisions\/8230"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=8229"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=8229"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=8229"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}