{"id":7243,"date":"2025-08-13T04:42:00","date_gmt":"2025-08-12T19:42:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=7243"},"modified":"2025-08-03T16:47:51","modified_gmt":"2025-08-03T07:47:51","slug":"mmbench-gui-hierarchical-multi-platform-evaluation-framework-for-gui-agents","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=7243","title":{"rendered":"MMBench-GUI: Hierarchical Multi-Platform Evaluation Framework for GUI Agents"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>MMBench-GUI: Hierarchical Multi-Platform Evaluation Framework for GUI Agents\u00a0<\/strong>[88.4]<br>MMBench-GUI\u306f\u3001Windows\u3001Linux\u3001iOS\u3001Android\u3001Web\u30d7\u30e9\u30c3\u30c8\u30d5\u30a9\u30fc\u30e0\u3067GUI\u81ea\u52d5\u5316\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092\u8a55\u4fa1\u3059\u308b\u968e\u5c64\u7684\u306a\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u3042\u308b\u3002 GUI\u30b3\u30f3\u30c6\u30f3\u30c4\u7406\u89e3\u3001\u8981\u7d20\u30b0\u30e9\u30a6\u30f3\u30c7\u30a3\u30f3\u30b0\u3001\u30bf\u30b9\u30af\u81ea\u52d5\u5316\u3001\u30bf\u30b9\u30af\u30b3\u30e9\u30dc\u30ec\u30fc\u30b7\u30e7\u30f3\u306e4\u3064\u306e\u30ec\u30d9\u30eb\u3067\u69cb\u6210\u3055\u308c\u3066\u304a\u308a\u3001GUI\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306b\u5fc5\u8981\u306a\u30b9\u30ad\u30eb\u3092\u30ab\u30d0\u30fc\u3057\u3066\u3044\u307e\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2507.19478v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2507.19478v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Fri, 25 Jul 2025 17:59:26 GMT)<\/li>\n\n\n\n<li>GUI\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u8a55\u4fa1\u7528\u306e\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3002\u300c(1) GUI Content Understanding, (2) GUI Element Grounding, (3) GUI Task Automation, and (4) GUI Task Collaboration.\u300d\u306e4\u6bb5\u968e\u3002\u300cFinding 1: General-purpose language models excel at task decomposition, planning, and self-reflection but struggle with fine-grained visual interactions.\u300d\u3001\u300cFinding 2: Accurate visual grounding significantly determines the success rate of GUI task execution.\u300d\u306f\u73fe\u5728\u306eGUI\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u958b\u767a\u306e\u65b9\u5411\u6027\u3068\u3082\u5408\u81f4\u3057\u3066\u3044\u308b\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/open-compass\/MMBench-GUI\">open-compass\/MMBench-GUI: Official repo of &#8220;MMBench-GUI: Hierarchical Multi-Platform Evaluation Framework for GUI Agents&#8221;. It can be used to evaluate a GUI agent with a hierarchical manner across multiple platforms, including Windows, Linux, macOS, iOS, Android and Web.<\/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":[181,517],"class_list":["post-7243","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-gui-agent","tag-517"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7243","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=7243"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7243\/revisions"}],"predecessor-version":[{"id":7244,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7243\/revisions\/7244"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7243"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7243"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7243"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}