{"id":5244,"date":"2024-07-29T03:27:00","date_gmt":"2024-07-28T18:27:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5244"},"modified":"2024-07-29T03:27:00","modified_gmt":"2024-07-28T18:27:00","slug":"minitron-compact-language-models-via-pruning-and-knowledge-distillation","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5244","title":{"rendered":"MINITRON \/ Compact Language Models via Pruning and Knowledge Distillation"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Compact Language Models via Pruning and Knowledge Distillation\u00a0<\/strong>[61.6]<br>\u30df\u30cb\u30c8\u30ed\u30f3\u30e2\u30c7\u30eb\u3067\u306f\u3001\u30b9\u30af\u30e9\u30c3\u30c1\u304b\u3089\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u306b\u6bd4\u3079\u3066MMLU\u30b9\u30b3\u30a2\u304c\u6700\u592716%\u6539\u5584\u3057\u3066\u3044\u308b\u3002 \u3059\u3067\u306b\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u6e08\u307f\u306e15B\u30e2\u30c7\u30eb\u304b\u30898B\u30684B\u30e2\u30c7\u30eb\u3092\u62bd\u51fa\u3059\u308b\u306b\u306f\u3001\u30b9\u30af\u30e9\u30c3\u30c1\u304b\u3089\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3059\u308b\u3088\u308a\u3082\u3001\u30e2\u30c7\u30eb\u6bce\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30c8\u30fc\u30af\u30f3\u3092\u6700\u592740\u500d\u5c11\u306a\u304f\u3059\u308b\u5fc5\u8981\u304c\u3042\u308a\u307e\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2407.14679v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2407.14679v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Fri, 19 Jul 2024 21:47:57 GMT)<\/li>\n\n\n\n<li>Nemotron 15B\u304b\u3089\u5f97\u3089\u308c\u305f\u9ad8\u6027\u80fd\u306a8B\u30e2\u30c7\u30eb\u53ca\u30734B\u30e2\u30c7\u30eb\u3002pruning\u3068distillation\u3092\u7d44\u307f\u5408\u308f\u305b\u305f\u30d9\u30b9\u30c8\u30d7\u30e9\u30af\u30c6\u30a3\u30b9\u3092\u5831\u544a\u3002<a href=\"https:\/\/devneko.jp\/wordpress\/?p=5055\">Gemma2, CriticGPT \u2013 arXiv\u6700\u65b0\u8ad6\u6587\u306e\u7d39\u4ecb (devneko.jp)<\/a>\u306e\u3068\u304d\u3082\u84b8\u7559\u304c\u7528\u3044\u3089\u308c\u3066\u3044\u305f\u304c\u3001\u5927\u898f\u6a21\u306a\u30e2\u30c7\u30eb\u304b\u3089\u5c0f\u898f\u6a21\u9ad8\u6027\u80fd\u306a\u30e2\u30c7\u30eb\u3092\u5f97\u308b\u3088\u3046\u306a\u624b\u9806\u304c\u4e00\u822c\u7684\u306b\u306a\u308b\u306e\u3060\u308d\u3046\u304b\u30fb\u30fb\u30fb<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/NVlabs\/Minitron\">GitHub &#8211; NVlabs\/Minitron: A family of compressed models obtained via pruning and knowledge distillation<\/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":[223,317,635],"class_list":["post-5244","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-pruning","tag-635"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5244","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=5244"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5244\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5244"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5244"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5244"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}