{"id":7990,"date":"2025-12-29T05:08:00","date_gmt":"2025-12-28T20:08:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=7990"},"modified":"2025-12-27T15:13:38","modified_gmt":"2025-12-27T06:13:38","slug":"toward-training-superintelligent-software-agents-through-self-play-swe-rl","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=7990","title":{"rendered":"Toward Training Superintelligent Software Agents through Self-Play SWE-RL"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Toward Training Superintelligent Software Agents through Self-Play SWE-RL\u00a0<\/strong>[66.1]<br>\u30bb\u30eb\u30d5\u30d7\u30ec\u30a4SWE-RL\u306f\u3001\u8d85\u77e5\u80fd\u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30d1\u30e9\u30c0\u30a4\u30e0\u306b\u5411\u3051\u305f\u7b2c\u4e00\u6b69\u3067\u3042\u308b\u3002 \u5f53\u793e\u306e\u30a2\u30d7\u30ed\u30fc\u30c1\u3067\u306f,\u30bd\u30fc\u30b9\u30b3\u30fc\u30c9\u3068\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u6e08\u307f\u306e\u4f9d\u5b58\u95a2\u4fc2\u3092\u5099\u3048\u305f\u30b5\u30f3\u30c9\u30dc\u30c3\u30af\u30b9\u30ea\u30dd\u30b8\u30c8\u30ea\u3078\u306e\u30a2\u30af\u30bb\u30b9\u306e\u307f\u3092\u5fc5\u8981\u3068\u3057\u3066\u3044\u307e\u3059\u3002 \u6211\u3005\u306e\u6210\u679c\u306f\u3001\u65e9\u3044\u6bb5\u968e\u3067\u3001\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u304c\u73fe\u5b9f\u4e16\u754c\u306e\u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u30ea\u30dd\u30b8\u30c8\u30ea\u304b\u3089\u5e83\u7bc4\u56f2\u306b\u308f\u305f\u308b\u5b66\u7fd2\u7d4c\u9a13\u3092\u81ea\u5f8b\u7684\u306b\u53ce\u96c6\u3059\u308b\u9053\u306e\u308a\u3092\u793a\u5506\u3057\u3066\u3044\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2512.18552v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2512.18552v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Sun, 21 Dec 2025 00:49:40 GMT)<\/li>\n\n\n\n<li>\u300cThe core idea of Self-play SWE-RL (SSR) is to allow LLM agents to self-improve through an iterative cycle of solving self-generated bugs and creating more complex challenges. As shown in Figure 1, the same LLM policy is divided into two roles: a bug-injection agent and a bug-solving agent.\u300d\u3068\u5bfe\u6226\u578b\u306e\u81ea\u5df1\u6539\u5584\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3002<a href=\"https:\/\/github.com\/facebookresearch\/cwm\">GitHub &#8211; facebookresearch\/cwm: Research code artifacts for Code World Model (CWM) including inference tools, reproducibility, and documentation.<\/a>\u3092\u30d9\u30fc\u30b9\u30e2\u30c7\u30eb\u3068\u3057\u3066\u52b9\u679c\u3092\u78ba\u8a8d\u3068\u306e\u3053\u3068\u3002<\/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":[356],"class_list":["post-7990","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-self-x"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7990","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=7990"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7990\/revisions"}],"predecessor-version":[{"id":7991,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7990\/revisions\/7991"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7990"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7990"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7990"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}