{"id":6007,"date":"2025-01-07T04:16:00","date_gmt":"2025-01-06T19:16:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6007"},"modified":"2025-01-07T04:16:00","modified_gmt":"2025-01-06T19:16:00","slug":"training-software-engineering-agents-and-verifiers-with-swe-gym","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6007","title":{"rendered":"Training Software Engineering Agents and Verifiers with SWE-Gym\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Training Software Engineering Agents and Verifiers with SWE-Gym\u00a0<\/strong>[89.6]<br>SWE-Gym\u306f\u3001\u73fe\u5b9f\u4e16\u754c\u306e\u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u30a8\u30f3\u30b8\u30cb\u30a2\u30ea\u30f3\u30b0(SWE)\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3059\u308b\u305f\u3081\u306e\u6700\u521d\u306e\u74b0\u5883\u3067\u3042\u308b\u3002 SWE-Gym\u306b\u306f2,438\u306e\u73fe\u5b9f\u4e16\u754c\u306ePython\u30bf\u30b9\u30af\u30a4\u30f3\u30b9\u30bf\u30f3\u30b9\u304c\u542b\u307e\u308c\u3066\u3044\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2412.21139v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2412.21139v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Mon, 30 Dec 2024 18:15:39 GMT)<\/li>\n\n\n\n<li>\u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u30a8\u30f3\u30b8\u30cb\u30a2\u30ea\u30f3\u30b0\u7528\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u958b\u767a\u306e\u305f\u3081\u306e\u74b0\u5883\u306e\u63d0\u6848\u3001\u304a\u3088\u3073\u3001\u9ad8\u6027\u80fd\u306a\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u958b\u767a\u3002o3\u3067\u5727\u5012\u7684\u306a\u7d50\u679c\u3092\u898b\u305f\u5f8c\u3067\u306f\u3042\u308b\u304c\u3001\u300cThrough extensive experiments, we demonstrate that SWE-Gym enables both agent and verifier models to achieve significant improvements in resolving complex software tasks. Our findings highlight the scalability of these approaches, revealing potential for continuous performance gains with increased compute.\u300d\u3068\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u7684\u52d5\u4f5c\u306e\u6709\u52b9\u6027\u306f\u9ad8\u3044\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/SWE-Gym\/SWE-Gym\">GitHub &#8211; SWE-Gym\/SWE-Gym<\/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":[42,486],"class_list":["post-6007","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-autonomous-agent","tag-486"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6007","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=6007"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6007\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6007"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6007"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6007"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}