{"id":4905,"date":"2024-05-30T05:14:00","date_gmt":"2024-05-29T20:14:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4905"},"modified":"2024-05-30T05:14:00","modified_gmt":"2024-05-29T20:14:00","slug":"jiuzhang3-0-efficiently-improving-mathematical-reasoning-by-training-small-data-synthesis-models","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4905","title":{"rendered":"JiuZhang3.0: Efficiently Improving Mathematical Reasoning by Training Small Data Synthesis Models"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>JiuZhang3.0: Efficiently Improving Mathematical Reasoning by Training Small Data Synthesis Models\u00a0<\/strong>[110.5]<br>\u65e2\u5b58\u306e\u7814\u7a76\u306f\u3001\u4e8b\u524d\u5b66\u7fd2\u306e\u305f\u3081\u306e\u5927\u898f\u6a21\u306a\u6570\u5b66\u95a2\u9023\u306e\u30c6\u30ad\u30b9\u30c8\u3092\u53ce\u96c6\u3057\u305f\u308a\u3001\u5de8\u5927\u306a\u6570\u5b66\u554f\u984c\u3092\u5408\u6210\u3059\u308b\u305f\u3081\u306b\u5f37\u529b\u306aLLM\u306b\u4f9d\u5b58\u3057\u3066\u3044\u308b\u3002 \u305d\u3053\u3067\u672c\u7814\u7a76\u3067\u306f,\u6570\u5b66\u554f\u984c\u5408\u6210\u306e\u305f\u3081\u306e\u5c0f\u3055\u306aLLM\u3092\u52b9\u7387\u7684\u306b\u5b66\u7fd2\u3057,\u9ad8\u54c1\u8cea\u306a\u4e8b\u524d\u5b66\u7fd2\u30c7\u30fc\u30bf\u3092\u52b9\u7387\u7684\u306b\u751f\u6210\u3059\u308b\u624b\u6cd5\u3092\u63d0\u6848\u3059\u308b\u3002 \u6211\u3005\u306f\u3001GPT-4 API 9.3k\u56de\u306e\u547c\u3073\u51fa\u3057\u30684.6B\u30c7\u30fc\u30bf\u306e\u4e8b\u524d\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u306e\u307f\u3092\u5fc5\u8981\u3068\u3059\u308b\u3001JuZhang3.0\u30e2\u30c7\u30eb\u306e\u4e8b\u524d\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u306b600\u4e07\u306e\u6570\u5b66\u554f\u984c\u3092\u5408\u6210\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2405.14365v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2405.14365v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 23 May 2024 09:43:19 GMT)<\/li>\n\n\n\n<li>\u6570\u5b66\u554f\u984c\u306b\u3064\u3044\u3066\u9ad8\u54c1\u8cea\u306a\u5408\u6210\u30c7\u30fc\u30bf\u3092\u69cb\u7bc9\u3057\u5c0f\u898f\u6a21LLM\u3092finetuning\u3001\u512a\u308c\u305f\u6027\u80fd\u3092\u9054\u6210\u3068\u306e\u3053\u3068\u3002\u5546\u7528\u3067\u306f\u30e9\u30a4\u30bb\u30f3\u30b9\u7684\u306b\u96e3\u3057\u3044\u5834\u5408\u3082\u591a\u3044\u304c\u6709\u671b\u306a\u30a2\u30d7\u30ed\u30fc\u30c1\u3067\u3042\u308b\u3068\u601d\u3046\u3002\u5b9f\u9a13\u3067JiuZhang3.0-8B (LLaMA-3-8B\u30d9\u30fc\u30b9) \u30687B (Mistral-7B\u30d9\u30fc\u30b9\uff09\u3092\u69cb\u7bc9\u3057\u3066\u3044\u3066\u65e9\u901fLlama 3\u3092\u53d6\u308a\u8fbc\u3093\u3067\u3044\u308b\u306e\u304c\u3059\u3054\u3044\uff06\u4e21\u8005\u306e\u6027\u80fd\u304c\u30bf\u30b9\u30af\u306b\u3088\u3063\u3066\u5927\u304d\u304f\u7570\u306a\u308b\u306e\u304c\u8208\u5473\u6df1\u3044\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/RUCAIBox\/JiuZhang3.0\">GitHub &#8211; RUCAIBox\/JiuZhang3.0: The code and data for the paper JiuZhang3.0<\/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":[223,390],"class_list":["post-4905","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-synthetic-data"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4905","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=4905"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4905\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4905"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4905"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4905"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}