{"id":5260,"date":"2024-08-08T05:26:00","date_gmt":"2024-08-07T20:26:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5260"},"modified":"2024-08-08T05:26:00","modified_gmt":"2024-08-07T20:26:00","slug":"towards-effective-and-efficient-continual-pre-training-of-large-language-models","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5260","title":{"rendered":"Towards Effective and Efficient Continual Pre-training of Large Language Models\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Towards Effective and Efficient Continual Pre-training of Large Language Models\u00a0<\/strong>[163.3]<br>CPT(Continuous pre-training)\u306f\u3001\u7279\u5b9a\u306e\u30c9\u30e1\u30a4\u30f3\u3084\u30bf\u30b9\u30af\u306b\u8a00\u8a9e\u30e2\u30c7\u30eb\u3092\u9069\u7528\u3059\u308b\u4e0a\u3067\u91cd\u8981\u306a\u30a2\u30d7\u30ed\u30fc\u30c1\u3067\u3042\u308b\u3002 \u672c\u7a3f\u3067\u306f,Llama-3 (8B) \u306e\u7d99\u7d9a\u4e8b\u524d\u8a13\u7df4\u306b\u95a2\u3059\u308b\u6280\u8853\u7684\u5831\u544a\u3092\u5831\u544a\u3059\u308b\u3002 \u30d0\u30c3\u30af\u30dc\u30fc\u30f3\u30e2\u30c7\u30eb\u306e\u4e2d\u56fd\u8a9e\u80fd\u529b\u3068\u79d1\u5b66\u7684\u63a8\u8ad6\u80fd\u529b\u3092\u5927\u5e45\u306b\u5411\u4e0a\u3055\u305b\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2407.18743v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2407.18743v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Fri, 26 Jul 2024 13:55:21 GMT)<\/li>\n\n\n\n<li>Llama-3\u306b\u5bfe\u3057\u3066\u7d99\u7d9a\u5b66\u7fd2\u3092\u5b9f\u65bd\u3001\u4e2d\u56fd\u8a9e\u80fd\u529b\u3068\u79d1\u5b66\u6280\u8853\u95a2\u9023\u306e\u63a8\u8ad6\u80fd\u529b\u3092\u5f37\u5316\u3057\u305f\u5831\u544a\u3002\u5408\u6210\u30c7\u30fc\u30bf\u3092\u6d3b\u7528\u3057\u3066\u3044\u308b\u70b9\u304c\u8208\u5473\u6df1\u3044\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/RUC-GSAI\/Llama-3-SynE\">GitHub &#8211; RUC-GSAI\/Llama-3-SynE<\/a>\u3001\u73fe\u6642\u70b9\u3067\u306fcoming soon<\/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":[78,222,223,390],"class_list":["post-5260","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-continual-learning","tag-llama-3","tag-llm","tag-synthetic-data"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5260","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=5260"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5260\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5260"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5260"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5260"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}