{"id":6259,"date":"2025-02-17T05:11:00","date_gmt":"2025-02-16T20:11:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6259"},"modified":"2025-02-17T05:11:00","modified_gmt":"2025-02-16T20:11:00","slug":"an-open-recipe-adapting-language-specific-llms-to-a-reasoning-model-in-one-day-via-model-merging","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6259","title":{"rendered":"An Open Recipe: Adapting Language-Specific LLMs to a Reasoning Model in One Day via Model Merging"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>An Open Recipe: Adapting Language-Specific LLMs to a Reasoning Model in One Day via Model Merging\u00a0<\/strong>[12.1]<br>\u672c\u7a3f\u3067\u306f,\u8a00\u8a9e\u56fa\u6709\u306e\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb(LLM)\u306e\u63a8\u8ad6\u80fd\u529b\u306e\u5411\u4e0a\u3092\u76ee\u7684\u3068\u3059\u308b\u3002 DeepSeek R1\u306f\u63a8\u8ad6\u306b\u512a\u308c\u3066\u3044\u307e\u3059\u304c\u3001\u4e3b\u306b\u82f1\u8a9e\u3084\u4e2d\u56fd\u8a9e\u306e\u3088\u3046\u306a\u9ad8\u30ea\u30bd\u30fc\u30b9\u8a00\u8a9e\u306b\u30e1\u30ea\u30c3\u30c8\u304c\u3042\u308a\u307e\u3059\u3002 \u4f4e\u30ea\u30bd\u30fc\u30b9\u8a00\u8a9e\u306f\u3001\u82f1\u8a9e\u4e2d\u5fc3\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c7\u30fc\u30bf\u3068\u30e2\u30c7\u30eb\u6700\u9069\u5316\u306e\u512a\u4f4d\u6027\u306e\u305f\u3081\u3001\u3044\u307e\u3060\u306b\u4fdd\u5b58\u3055\u308c\u3066\u3044\u306a\u3044\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2502.09056v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2502.09056v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 13 Feb 2025 08:10:45 GMT)<\/li>\n\n\n\n<li>LLM\u306e\u63a8\u8ad6\u80fd\u529b\u3092\u9ad8\u3081\u308b\u305f\u3081\u306e\u30e2\u30c7\u30eb\u30de\u30fc\u30b8+SFT\u3001\u300cWe demonstrate that, with only publicly available datasets and a computational budget of $120, it is possible to enhance the reasoning capabilities of language-specific LLMs to match the level of DeepSeek R1, without compromising their performance on target language tasks.\u300d\u3068\u306e\u3053\u3068<\/li>\n\n\n\n<li><a href=\"https:\/\/rinna.co.jp\/news\/2025\/02\/20250213.html\">Qwen2.5\u3068DeepSeek R1\u3092\u5229\u7528\u3057\u305f\u65e5\u672c\u8a9e\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb\u300cQwen2.5 Bakeneko 32B\u300d\u30b7\u30ea\u30fc\u30ba\u3092\u516c\u958b\uff5crinna\u682a\u5f0f\u4f1a\u793e<\/a>\u3067\u3082\u8fd1\u3044\u30a2\u30d7\u30ed\u30fc\u30c1\u3092\u3068\u3063\u3066\u3044\u308b\u3088\u3046\u306b\u898b\u3048\u308b\u3002<\/li>\n\n\n\n<li><\/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":[232,256],"class_list":["post-6259","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-lrm","tag-model-merging"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6259","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=6259"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6259\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6259"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6259"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6259"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}