{"id":7823,"date":"2025-11-27T06:31:00","date_gmt":"2025-11-26T21:31:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=7823"},"modified":"2025-11-23T20:34:00","modified_gmt":"2025-11-23T11:34:00","slug":"agent0-unleashing-self-evolving-agents-from-zero-data-via-tool-integrated-reasoning","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=7823","title":{"rendered":"Agent0: Unleashing Self-Evolving Agents from Zero Data via Tool-Integrated Reasoning"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Agent0: Unleashing Self-Evolving Agents from Zero Data via Tool-Integrated Reasoning\u00a0<\/strong>[84.7]<br>\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb(LLM)\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306f\u3001\u4eba\u9593\u306e\u8a08\u7b97\u30c7\u30fc\u30bf\u3078\u306e\u4f9d\u5b58\u306b\u3088\u3063\u3066\u5236\u7d04\u3055\u308c\u308b\u3002 \u6211\u3005\u306f,\u5916\u90e8\u30c7\u30fc\u30bf\u3092\u6301\u305f\u306a\u3044\u9ad8\u6027\u80fd\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092\u9032\u5316\u3055\u305b\u308b\u5b8c\u5168\u81ea\u5f8b\u578b\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3067\u3042\u308bAgent0\u3092\u7d39\u4ecb\u3059\u308b\u3002 Agent0\u306f\u63a8\u8ad6\u80fd\u529b\u3092\u5927\u5e45\u306b\u5411\u4e0a\u3055\u305b\u3001Qwen3-8B-Base\u30e2\u30c7\u30eb\u3092\u6570\u5b66\u7684\u63a8\u8ad6\u306718%\u6539\u5584\u3057\u3001\u4e00\u822c\u7684\u306a\u63a8\u8ad6\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u306724%\u6539\u5584\u3057\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2511.16043v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2511.16043v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 20 Nov 2025 05:01:57 GMT)<\/li>\n\n\n\n<li>\u300cwe initialize two functionally distinct agents: an execu- tor agent and a curriculum agent. These agents co-evolve through a symbiotic competition: the curriculum agent is trained using RL (Shao et al , 2024) to propose frontier tasks that precisely challenge the executor\u2019s current capabilities, using the executor\u2019s uncertainty (i.e., self-consistency across multiple answers) and its frequency of tool use as reward signals. Concurrently, the executor agent is trained via RL to successfully solve these tasks, optimizing on a filtered set of challenging problems generated by the frozen curriculum agent and using pseudo-labels derived from its own majority voting. Equipping the executor with a tool enhances its problem-solving abilities, which in turn com- pels the tool-equipped curriculum agent to generate more complex, tool-based curricula.\u300d\u3068\u3044\u3046\u8907\u6570\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092\u6d3b\u7528\u3057\u305f\u5171\u9032\u5316\u306a\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3002Agent\u69cb\u7bc9\u306b\u304a\u3044\u3066\u3082\u8fd1\u3044\u30a2\u30d7\u30ed\u30fc\u30c1\u304c\u6d41\u884c\u3063\u3066\u3044\u308b\u3088\u3046\u306b\u601d\u3046\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/aiming-lab\/Agent0\">GitHub &#8211; aiming-lab\/Agent0: [arXiv&#8217;25] Agent0: Unleashing Self-Evolving Agents from Zero Data via Tool-Integrated Reasoning<\/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],"class_list":["post-7823","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-autonomous-agent"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7823","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=7823"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7823\/revisions"}],"predecessor-version":[{"id":7824,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7823\/revisions\/7824"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7823"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7823"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7823"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}