{"id":3148,"date":"2023-04-07T06:38:00","date_gmt":"2023-04-06T21:38:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=3148"},"modified":"2023-04-07T06:38:00","modified_gmt":"2023-04-06T21:38:00","slug":"llama-adapter","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=3148","title":{"rendered":"LLaMA-Adapter"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention\u00a0<\/strong>[53.8]<br>\u547d\u4ee4\u8ffd\u5f93\u30e2\u30c7\u30eb\u306bLLaMA\u3092\u52b9\u7387\u3088\u304f\u5fae\u8abf\u6574\u3059\u308b\u65b9\u6cd5\u3067\u3042\u308bLLaMA-Adapter\u3092\u63d0\u6848\u3059\u308b\u3002 LLaMA-Adapter\u306f52K\u306e\u81ea\u5df1\u547d\u4ee4\u578b\u30c7\u30e2\u3092\u4f7f\u7528\u3057\u3066\u3001\u51cd\u7d50\u3057\u305fLLaMA 7B\u30e2\u30c7\u30eb\u306b1.2M\u306e\u5b66\u7fd2\u53ef\u80fd\u306a\u30d1\u30e9\u30e1\u30fc\u30bf\u3092\u5c0e\u5165\u3057\u3066\u3044\u308b\u3002 \u52b9\u7387\u7684\u306a\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u306b\u3088\u308a\u3001LLaMA-Adapter\u306f\u3001\u5b8c\u5168\u306b\u5fae\u8abf\u6574\u3055\u308c\u305f7B\u30d1\u30e9\u30e1\u30fc\u30bf\u3092\u6301\u3064Alpaca\u306b\u5339\u6575\u3059\u308b\u9ad8\u54c1\u8cea\u306a\u5fdc\u7b54\u3092\u751f\u6210\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2303.16199v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2303.16199v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Tue, 28 Mar 2023 17:59:12 GMT)<\/li>\n\n\n\n<li>\u305d\u306e\u540d\u306e\u901a\u308aLLaMA\u3078\u306eAdapter\u30011.2M\u30d1\u30e9\u30e1\u30fc\u30bf\u3068\u6975\u3081\u3066\u5c11\u306a\u3044\u30d1\u30e9\u30e1\u30fc\u30bf\u306e\u30c1\u30e5\u30fc\u30cb\u30f3\u30b0\u306e\u307f\u3067\u5b8c\u5168\u306afine tuning\u306b\u76f8\u5f53\u3068\u306e\u3053\u3068<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/ZrrSkywalker\/LLaMA-Adapter\">GitHub &#8211; ZrrSkywalker\/LLaMA-Adapter: Fine-tuning LLaMA to follow instructions within 1 Hour and 1.2M Parameters<\/a><\/li>\n<\/ul>\n\n\n\n<p>LLM\u306e\u6642\u4ee3\u306b\u52b9\u7387\u306e\u3088\u3044\u8abf\u6574\u306f\u3069\u3046\u3042\u308b\u3079\u304d\u304b\u306f\u3068\u3066\u3082\u5927\u4e8b\u3001\u3060\u304cweight\u3092\u5909\u66f4\u53ef\u80fd\u304b\u306f\u60a9\u307e\u3057\u3044\u6642\u4ee3\u304b\u3082\u3068\u3082\u601d\u3046\u3002<a href=\"https:\/\/github.com\/txsun1997\/Black-Box-Tuning\">GitHub &#8211; txsun1997\/Black-Box-Tuning: ICML&#8217;2022: Black-Box Tuning for Language-Model-as-a-Service &amp; EMNLP&#8217;2022: BBTv2: Towards a Gradient-Free Future with Large Language Models<\/a>\u3068\u304b\u306b\u3082\u6ce8\u76ee\u3057\u3066\u3044\u305f\u308a\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>LLM\u306e\u6642\u4ee3\u306b\u52b9\u7387\u306e\u3088\u3044\u8abf\u6574\u306f\u3069\u3046\u3042\u308b\u3079\u304d\u304b\u306f\u3068\u3066\u3082\u5927\u4e8b\u3001\u3060\u304cweight\u3092\u5909\u66f4\u53ef\u80fd\u304b\u306f\u60a9\u307e\u3057\u3044\u6642\u4ee3\u304b\u3082\u3068\u3082\u601d\u3046\u3002GitHub &#8211; txsun1997\/Black-Box-Tuning: ICML&#038;#8217 &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=3148\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;LLaMA-Adapter&#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":[10],"class_list":["post-3148","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-adapter"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/3148","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=3148"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/3148\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3148"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3148"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3148"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}