{"id":6459,"date":"2025-03-19T03:56:00","date_gmt":"2025-03-18T18:56:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6459"},"modified":"2025-03-19T03:56:00","modified_gmt":"2025-03-18T18:56:00","slug":"search-r1-training-llms-to-reason-and-leverage-search-engines-with-reinforcement-learning","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6459","title":{"rendered":"Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning\u00a0<\/strong>[45.7]<br>\u672c\u7a3f\u3067\u306f,\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb(LLM)\u306e\u305f\u3081\u306eDeepSeek-R1\u30e2\u30c7\u30eb\u306e\u62e1\u5f35\u3067\u3042\u308bSearch-R1\u3092\u7d39\u4ecb\u3059\u308b\u3002 Search-R1\u306f\u3001\u30ea\u30a2\u30eb\u30bf\u30a4\u30e0\u691c\u7d22\u306b\u3088\u308b\u30b9\u30c6\u30c3\u30d7\u30d0\u30a4\u30b9\u30c6\u30c3\u30d7\u63a8\u8ad6\u4e2d\u306b(\u8907\u6570\u306e)\u691c\u7d22\u30af\u30a8\u30ea\u3092\u81ea\u5f8b\u7684\u306b\u751f\u6210\u3059\u308b\u3002 \u5b9f\u9a13\u306e\u7d50\u679c\u3001\u30b5\u30fc\u30c1-R1\u306f26%(Qwen2.5-7B)\u300121%(Qwen2.5-3B)\u300110%(LLaMA3.2-3B)\u306eSOTA\u30d9\u30fc\u30b9\u30e9\u30a4\u30f3\u306e\u6027\u80fd\u5411\u4e0a\u3092\u793a\u3057\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2503.09516v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2503.09516v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 12 Mar 2025 16:26:39 GMT)<\/li>\n\n\n\n<li>\u691c\u7d22\u30af\u30a8\u30ea\u3092\u767a\u884c\u3057\u306a\u304c\u3089\u63a8\u8ad6\u3092\u9032\u3081\u308b\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u306e\u63d0\u6848\u300cSEARCH-R1, a novel reinforcement learning framework that enables large language models (LLMs) to interleave self-reasoning with real-time search engine interactions.\u300d\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/PeterGriffinJin\/Search-R1\">GitHub &#8211; PeterGriffinJin\/Search-R1: Search-R1: An Efficient, Scalable RL Training Framework for Reasoning &amp; Search Engine Calling interleaved LLM based on veRL<\/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":[324,600],"class_list":["post-6459","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-rag","tag-600"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6459","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=6459"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6459\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6459"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6459"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6459"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}