{"id":2475,"date":"2022-10-24T04:05:00","date_gmt":"2022-10-23T19:05:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=2475"},"modified":"2022-10-24T04:05:00","modified_gmt":"2022-10-23T19:05:00","slug":"scaling-instruction-finetuned-language-models","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=2475","title":{"rendered":"Scaling Instruction-Finetuned Language Models"},"content":{"rendered":"\n<ul class=\"wp-block-list\"><li><strong>Scaling Instruction-Finetuned Language Models\u00a0<\/strong>[126.5]<br>\u547d\u4ee4\u3068\u3057\u3066\u8868\u73fe\u3055\u308c\u305f\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306e\u96c6\u5408\u4e0a\u3067\u306e\u8a00\u8a9e\u30e2\u30c7\u30eb\u306e\u5fae\u8abf\u6574\u306f\u3001\u30e2\u30c7\u30eb\u6027\u80fd\u3092\u6539\u5584\u3059\u308b\u3053\u3068\u304c\u793a\u3055\u308c\u3066\u3044\u308b\u3002 \u547d\u4ee4\u306e\u5fae\u8abf\u6574\u306b\u3088\u308a,\u69d8\u3005\u306a\u30e2\u30c7\u30eb\u30af\u30e9\u30b9\u306e\u6027\u80fd\u304c\u5287\u7684\u306b\u5411\u4e0a\u3059\u308b\u3053\u3068\u304c\u308f\u304b\u3063\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2210.11416v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2210.11416v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 20 Oct 2022 16:58:32 GMT)<ul><li>1,836\u306e\u30bf\u30b9\u30af\u3067fine tuning\uff08instruction finetuning\uff09\u3059\u308b\u3053\u3068\u3067\u69d8\u3005\u306a\u30bf\u30b9\u30af\u306b\u5bfe\u3059\u308b\u6027\u80fd\u304c\u5411\u4e0a\u3057\u305f\u3068\u3059\u308b\u8ad6\u6587\u3002PaLM\u306b\u5bfe\u3057\u3066instruction fine tuning\u3092\u884c\u3063\u305fFlan-PaLM 540B\u306f\u3082\u3068\u3082\u3068\u306ePaLM\u3092\u5927\u5e45\u306b\u6027\u80fd\u6539\u5584\u3057\u3001five-shot MMLU\u3067SoTA\u3002\u4f7f\u7528\u3059\u308b\u30bf\u30b9\u30af\u30b5\u30a4\u30ba\u304c\u5897\u3048\u308b\u307b\u3069\u6027\u80fd\u5411\u4e0a\u50be\u5411\u304c\u307f\u3089\u308c\u308b\u304c\u3001282\u30bf\u30b9\u30af\u3067\u6982\u306d\u982d\u6253\u3061\u306b\u306a\u3063\u3066\u3044\u308b<ul><li>\u305d\u308c\u4ee5\u4e0a\u5897\u3084\u3057\u3066\u3082\u65b0\u305f\u306a\u77e5\u8b58\u3092\u63d0\u4f9b\u3057\u3066\u3044\u306a\u3044\u306e\u3067\u306f\uff1f\u3082\u3057\u304f\u306f\u8a00\u8a9e\u30e2\u30c7\u30eb\u5185\u306e\u77e5\u8b58\u3092\u8868\u73fe\u3059\u308b\u305f\u3081\u306b\u4e00\u5b9a\u4ee5\u4e0a\u306e\u30bf\u30b9\u30af\u306f\u5f79\u306b\u7acb\u305f\u306a\u3044\u306e\u3067\u306f\uff1f\u3068\u306e\u6307\u6458\u304c\u3042\u308b\u304c\u3001\u65b0\u305f\u306a\u77e5\u8b58\u30fb\u60c5\u5831\u3092\u63d0\u4f9b\u3059\u308b\u305f\u3081\u306e\u6761\u4ef6\u306b\u3081\u3063\u3061\u3083\u8208\u5473\u304c\u3042\u308b\u3002<\/li><\/ul><\/li><\/ul><ul><li>\u4e0b\u8a18\u306bFlan-T5\u306e\u30c1\u30a7\u30c3\u30af\u30dd\u30a4\u30f3\u30c8\u304c\u516c\u958b\u3055\u308c\u3066\u3044\u308b\u3002PaLM 62B\u3088\u308afew-shot\u6027\u80fd\u304c\u512a\u308c\u3066\u3044\u308b\u3068\u306e\u3053\u3068\u3002<ul><li><a href=\"https:\/\/github.com\/google-research\/t5x\/blob\/main\/docs\/models.md#flan-t5-checkpoints\">t5x\/models.md at main \u00b7 google-research\/t5x (github.com)<\/a><\/li><\/ul><\/li><\/ul><\/li><\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Scaling Instruction-Finetuned Language Models\u00a0[126.5]\u547d\u4ee4\u3068\u3057\u3066\u8868\u73fe\u3055\u308c\u305f\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306e\u96c6\u5408\u4e0a\u3067\u306e\u8a00\u8a9e\u30e2\u30c7\u30eb\u306e\u5fae\u8abf\u6574\u306f\u3001\u30e2\u30c7\u30eb\u6027\u80fd\u3092\u6539\u5584\u3059\u308b\u3053\u3068\u304c\u793a\u3055\u308c\u3066\u3044\u308b\u3002 \u547d\u4ee4\u306e &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=2475\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;Scaling Instruction-Finetuned Language Models&#8221; \u306e<\/span>\u7d9a\u304d\u3092\u8aad\u3080<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[146,170,203],"class_list":["post-2475","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-fewshot","tag-gigantic-language-model","tag-instruction-finetuning"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/2475","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=2475"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/2475\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2475"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2475"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2475"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}