{"id":8184,"date":"2026-02-13T05:07:00","date_gmt":"2026-02-12T20:07:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=8184"},"modified":"2026-02-01T07:11:45","modified_gmt":"2026-01-31T22:11:45","slug":"os-marathon-benchmarking-computer-use-agents-on-long-horizon-repetitive-tasks","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=8184","title":{"rendered":"OS-Marathon: Benchmarking Computer-Use Agents on Long-Horizon Repetitive Tasks"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>OS-Marathon: Benchmarking Computer-Use Agents on Long-Horizon Repetitive Tasks\u00a0<\/strong>[37.0]<br>\u30ed\u30f3\u30b0\u30db\u30e9\u30a4\u30ba\u30f3\u3067\u53cd\u5fa9\u7684\u306a\u30bf\u30b9\u30af\u306f\u3001\u30d7\u30ed\u30d5\u30a7\u30c3\u30b7\u30e7\u30ca\u30eb\u306a\u8a2d\u5b9a\u3067\u4e00\u822c\u7684\u3067\u3042\u308b\u3002 \u3053\u308c\u3089\u306e\u30bf\u30b9\u30af\u306f\u3001\u51e6\u7406\u3059\u308b\u30c7\u30fc\u30bf\u306e\u30b5\u30a4\u30ba\u306b\u6bd4\u4f8b\u3057\u3066\u6975\u7aef\u306a\u9577\u3055\u307e\u3067\u62e1\u5f35\u3067\u304d\u308b\u305f\u3081\u3001\u4eba\u9593\u306b\u3068\u3063\u3066\u9000\u5c48\u306a\u4f5c\u696d\u3067\u3042\u308b\u3053\u3068\u304c\u591a\u3044\u3002 \u6211\u3005\u306f2\u3064\u306e\u30c9\u30e1\u30a4\u30f3\u306b\u307e\u305f\u304c\u308b242\u306e\u9577\u671f\u7684\u53cd\u5fa9\u7684\u306a\u30bf\u30b9\u30af\u304b\u3089\u306a\u308bOS-Marathon\u3092\u69cb\u7bc9\u3057\u3001SOTA(State-of-the-art)\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092\u8a55\u4fa1\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2601.20650v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2601.20650v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 28 Jan 2026 14:35:23 GMT)<\/li>\n\n\n\n<li>\u300cOS-Marathon is specifically tailored to evaluate CUA performance in long- horizon, repetitive execution scenarios, comprising 242 tasks across 2 domains and 7 distinct execution environments. \u300d\u3068\u9577\u671f\u304b\u3064\u53cd\u5fa9\u7684\u306a\u30bf\u30b9\u30af\u304c\u3042\u308bGUI\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3002\u304b\u306a\u308a\u96e3\u3057\u3044\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u306b\u898b\u3048\u308b\u3002<\/li>\n\n\n\n<li>\u30d7\u30ed\u30b8\u30a7\u30af\u30c8\u30b5\u30a4\u30c8\u306f<a href=\"https:\/\/os-marathon.github.io\/\">OS-Marathon Benchmark<\/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-8184","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\/8184","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=8184"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/8184\/revisions"}],"predecessor-version":[{"id":8185,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/8184\/revisions\/8185"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=8184"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=8184"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=8184"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}