{"id":8029,"date":"2026-01-06T06:27:00","date_gmt":"2026-01-05T21:27:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=8029"},"modified":"2026-01-04T11:09:11","modified_gmt":"2026-01-04T02:09:11","slug":"mai-ui-technical-report-real-world-centric-foundation-gui-agents","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=8029","title":{"rendered":"MAI-UI Technical Report: Real-World Centric Foundation GUI Agents\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>MAI-UI Technical Report: Real-World Centric Foundation GUI Agents\u00a0<\/strong>[33.5]<br>MAI-UI\u306f\u30012B\u30018B\u300132B\u3001\u304a\u3088\u3073235B-A22B\u3092\u542b\u3080\u5168\u7bc4\u56f2\u306eGUI\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u30d5\u30a1\u30df\u30ea\u30fc\u3067\u3042\u308b\u3002 \u30cd\u30a4\u30c6\u30a3\u30d6\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8-\u30e6\u30fc\u30b6\u30a4\u30f3\u30bf\u30e9\u30af\u30b7\u30e7\u30f3\u306e\u6b20\u5982\u3001UI\u306e\u307f\u306e\u64cd\u4f5c\u306e\u9650\u754c\u3001\u5b9f\u7528\u7684\u306a\u30c7\u30d7\u30ed\u30a4\u30e1\u30f3\u30c8\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u306e\u6b20\u5982\u3067\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2512.22047v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2512.22047v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Fri, 26 Dec 2025 14:51:52 GMT)<\/li>\n\n\n\n<li>\u300cMAI-UI establishes new state-of-the-art across GUI grounding and mobile navigation. On grounding benchmarks, it reaches 73.5% on ScreenSpot-Pro, 91.3% on MMBench GUI L2, 70.9% on OSWorld-G, and 49.2% on UI-Vision, surpassing Gemini-3-Pro and Seed1.8 on ScreenSpot-Pro.\u300d\u306a\u3069SoTA\u3092\u4e3b\u5f35\u3002\u5b9f\u7528\u30ec\u30d9\u30eb\u306b\u9054\u3057\u3064\u3064\u3042\u308b\u5370\u8c61\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/Tongyi-MAI\/MAI-UI\">GitHub &#8211; Tongyi-MAI\/MAI-UI: MAI-UI: Real-World Centric Foundation 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],"class_list":["post-8029","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-gui-agent"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/8029","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=8029"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/8029\/revisions"}],"predecessor-version":[{"id":8030,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/8029\/revisions\/8030"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=8029"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=8029"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=8029"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}