{"id":4010,"date":"2024-07-02T09:49:01","date_gmt":"2024-07-02T01:49:01","guid":{"rendered":"https:\/\/nullthought.net\/?p=4010"},"modified":"2024-11-07T11:05:37","modified_gmt":"2024-11-07T03:05:37","slug":"%e8%85%be%e8%ae%afai%e5%ae%9e%e9%aa%8c%e5%ae%a4%ef%bc%9a%e5%88%a9%e7%94%a810%e4%ba%bf%e4%b8%aa%e8%99%9a%e6%8b%9f%e8%ba%ab%e4%bb%bd%e6%89%a9%e5%b1%95%e5%90%88%e6%88%90%e6%95%b0%e6%8d%ae%e5%88%9b","status":"publish","type":"post","link":"https:\/\/nullthought.net\/?p=4010","title":{"rendered":"\u817e\u8bafAI\u5b9e\u9a8c\u5ba4\uff1a\u5229\u752810\u4ebf\u4e2a\u865a\u62df\u8eab\u4efd\u6269\u5c55\u5408\u6210\u6570\u636e\u521b\u5efa"},"content":{"rendered":"\n<p><a href=\"https:\/\/ai.tencent.com\/ailab\/en\/index\" target=\"_blank\" rel=\"noreferrer noopener\">\u817e\u8bafAI\u5b9e\u9a8c\u5ba4<\/a>\u8fd1\u65e5\u53d1\u8868\u8bba\u6587\u300a\u5229\u752810\u4ebf\u4e2a\u865a\u62df\u8eab\u4efd\u6269\u5c55\u5408\u6210\u6570\u636e\u521b\u5efa\u300b\uff08<a href=\"https:\/\/huggingface.co\/papers\/2406.20094\" target=\"_blank\" rel=\"noreferrer noopener\">Scaling Synthetic Data Creation with 1,000,000,000 Personas<\/a>\uff09\u3002\u8bba\u6587\u9488\u5bf9<a href=\"https:\/\/nullthought.net\/?p=3963\" target=\"_blank\" rel=\"noreferrer noopener\">\u5408\u6210\u6570\u636e<\/a>\u63d0\u51fa\u4e86\u5f88\u521b\u65b0\u7684\u601d\u8def\u548c\u6280\u672f\u65b9\u6cd5\ud83d\udc4d<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"755\" height=\"465\" src=\"https:\/\/nullthought.net\/wp-content\/uploads\/2024\/07\/image-1.png\" alt=\"\" class=\"wp-image-4011\" srcset=\"https:\/\/nullthought.net\/wp-content\/uploads\/2024\/07\/image-1.png 755w, https:\/\/nullthought.net\/wp-content\/uploads\/2024\/07\/image-1-300x185.png 300w\" sizes=\"auto, (max-width: 755px) 100vw, 755px\" \/><figcaption class=\"wp-element-caption\"><a href=\"https:\/\/huggingface.co\/papers\/2406.20094\" target=\"_blank\" rel=\"noreferrer noopener\">Scaling Synthetic Data Creation with 1,000,000,000 Personas<\/a><br><a href=\"https:\/\/ai.tencent.com\/ailab\/en\/index\" target=\"_blank\" rel=\"noreferrer noopener\">\u817e\u8bafAI\u5b9e\u9a8c\u5ba4\uff08Tencent AI Lab\uff09<\/a><\/figcaption><\/figure>\n\n\n\n<p>\u8bba\u6587\u8ba8\u8bba\u4e86\u4e00\u79cd\u5229\u7528\u5927\u578b\u8bed\u8a00\u6a21\u578b\uff08LLM\uff09\u8fdb\u884c\u5408\u6210\u6570\u636e\u521b\u5efa\u7684\u65b0\u65b9\u6cd5\u3002\u8be5\u65b9\u6cd5\u901a\u8fc7\u4ece\u7f51\u7edc\u6570\u636e\u4e2d\u6574\u7406\u51fa10\u4ebf\u4e2a\u591a\u6837\u5316\u7684\u865a\u62df\u8eab\u4efd\uff08Persona\uff09\u96c6\u5408\uff0c\u79f0\u4e3aPersona Hub\uff0c\u6765\u751f\u6210\u8de8\u5404\u79cd\u573a\u666f\u7684\u5408\u6210\u6570\u636e\u3002\u8fd9\u4e9b\u865a\u62df\u8eab\u4efd\u4f5c\u4e3a\u5206\u5e03\u5f0f\u7684\u4e16\u754c\u77e5\u8bc6\u8f7d\u4f53\uff0c\u4fc3\u8fdb\u4e86\u591a\u6837\u5316\u5408\u6210\u6570\u636e\u7684\u5927\u89c4\u6a21\u521b\u5efa\u3002<\/p>\n\n\n\n<h4 class=\"wp-block-heading\">\u4e00\u3001\u8bba\u6587\u5185\u5bb9\u5173\u952e\u70b9<\/h4>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Persona Hub<\/strong>:\n<ul class=\"wp-block-list\">\n<li>\u4e00\u4e2a\u5305\u542b10\u4ebf\u4e2a\u865a\u62df\u8eab\u4efd\u7684\u5e93\uff0c\u6bcf\u4e2a\u8eab\u4efd\u4ee3\u8868\u4e00\u4e2a\u72ec\u7279\u7684\u89c6\u89d2\u3002<\/li>\n\n\n\n<li>\u8fd9\u4e9b\u865a\u62df\u8eab\u4efd\u6765\u81ea\u516c\u5171\u7f51\u7edc\u6587\u672c\uff0c\u4ee3\u8868\u4e16\u754c\u77e5\u8bc6\u7684\u538b\u7f29\u5f62\u5f0f\u3002<\/li>\n\n\n\n<li>\u4f7f\u7528\u4e24\u79cd\u65b9\u6cd5\u751f\u6210\u865a\u62df\u8eab\u4efd\uff1a\u6587\u672c\u5230\u865a\u62df\u8eab\u4efd\uff08Text-to-Persona\uff09\u548c\u865a\u62df\u8eab\u4efd\u5230\u865a\u62df\u8eab\u4efd\uff08Persona-to-Persona\uff09\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u6587\u672c\u5230\u865a\u62df\u8eab\u4efd\uff08Text-to-Persona\uff09<\/strong>:\n<ul class=\"wp-block-list\">\n<li>\u901a\u8fc7\u5411LLM\u63d0\u793a\u7f51\u7edc\u6587\u672c\u6570\u636e\u751f\u6210\u865a\u62df\u8eab\u4efd\u3002<\/li>\n\n\n\n<li>\u5141\u8bb8\u63a7\u5236\u865a\u62df\u8eab\u4efd\u63cf\u8ff0\u7684\u7c92\u5ea6\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u865a\u62df\u8eab\u4efd\u5230\u865a\u62df\u8eab\u4efd\uff08Persona-to-Persona\uff09<\/strong>:\n<ul class=\"wp-block-list\">\n<li>\u901a\u8fc7\u4eba\u9645\u5173\u7cfb\u6d3e\u751f\u51fa\u65b0\u7684\u865a\u62df\u8eab\u4efd\uff0c\u589e\u5f3a\u591a\u6837\u6027\u3002<\/li>\n\n\n\n<li>\u4f7f\u7528\u516d\u5ea6\u5206\u9694\u7406\u8bba\u6269\u5c55\u865a\u62df\u8eab\u4efd\u5173\u7cfb\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u53bb\u91cd<\/strong>:\n<ul class=\"wp-block-list\">\n<li>\u4f7f\u7528MinHash\u548c\u57fa\u4e8e\u5d4c\u5165\u7684\u53bb\u91cd\u65b9\u6cd5\u786e\u4fdd\u591a\u6837\u6027\u3002<\/li>\n\n\n\n<li>\u6700\u7ec8\u96c6\u5408\u7531\u8d85\u8fc710\u4ebf\u4e2a\u72ec\u7279\u7684\u865a\u62df\u8eab\u4efd\u7ec4\u6210\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u6570\u636e\u5408\u6210<\/strong>:\n<ul class=\"wp-block-list\">\n<li>\u5c06\u865a\u62df\u8eab\u4efd\u96c6\u6210\u5230\u6570\u636e\u5408\u6210\u63d0\u793a\u4e2d\uff0c\u5f15\u5bfcLLM\u521b\u5efa\u5408\u6210\u6570\u636e\u3002<\/li>\n\n\n\n<li>\u5c55\u793a\u4e86\u5728\u521b\u5efa\u6570\u5b66\u95ee\u9898\u3001\u903b\u8f91\u63a8\u7406\u95ee\u9898\u3001\u6307\u4ee4\u3001\u77e5\u8bc6\u4e30\u5bcc\u7684\u6587\u672c\u3001\u6e38\u620fNPC\u548c\u5de5\u5177\u65b9\u9762\u7684\u591a\u529f\u80fd\u6027\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u4f7f\u7528\u6848\u4f8b<\/strong>:\n<ul class=\"wp-block-list\">\n<li>\u6570\u5b66\u95ee\u9898\uff1a\u5c55\u793a\u4e86\u521b\u5efa\u591a\u6837\u4e14\u5177\u6709\u6311\u6218\u6027\u7684\u6570\u5b66\u95ee\u9898\u3002<\/li>\n\n\n\n<li>\u903b\u8f91\u63a8\u7406\uff1a\u5448\u73b0\u4e86\u5178\u578b\u7684\u903b\u8f91\u63a8\u7406\u95ee\u9898\u548cRuozhiba\u98ce\u683c\u7684\u95ee\u9898\u3002<\/li>\n\n\n\n<li>\u6307\u4ee4\uff1a\u6a21\u62df\u7528\u6237\u63d0\u793a\uff0c\u589e\u5f3aLLM\u7684\u6307\u4ee4\u8ddf\u968f\u80fd\u529b\u3002<\/li>\n\n\n\n<li>\u77e5\u8bc6\u4e30\u5bcc\u7684\u6587\u672c\uff1a\u901a\u8fc7\u865a\u62df\u8eab\u4efd\u9a71\u52a8\u7684\u63d0\u793a\u751f\u6210\u4fe1\u606f\u4e30\u5bcc\u7684\u5185\u5bb9\u3002<\/li>\n\n\n\n<li>\u6e38\u620fNPC\uff1a\u5c06\u771f\u5b9e\u4e16\u754c\u7684\u865a\u62df\u8eab\u4efd\u6295\u5c04\u5230\u6e38\u620f\u89d2\u8272\u4e2d\uff0c\u521b\u5efa\u591a\u6837\u7684NPC\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u8bc4\u4f30<\/strong>:\n<ul class=\"wp-block-list\">\n<li>\u4f7f\u7528\u5408\u6210\u6d4b\u8bd5\u96c6\u8bc4\u4f30\u751f\u6210\u6570\u636e\u7684\u6709\u6548\u6027\u3002<\/li>\n\n\n\n<li>\u5fae\u8c03\u6a21\u578b\u5728MATH\u548c\u5408\u6210\u6d4b\u8bd5\u96c6\u4e0a\u53d6\u5f97\u4e86\u4ee4\u4eba\u5370\u8c61\u6df1\u523b\u7684\u51c6\u786e\u6027\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n\n\n\n<h4 class=\"wp-block-heading\">\u4e8c\u3001\u7ed3\u8bba<\/h4>\n\n\n\n<p>\u865a\u62df\u8eab\u4efd\u9a71\u52a8\u7684\u6570\u636e\u5408\u6210\u65b9\u6cd5\u5728\u5927\u89c4\u6a21\u751f\u6210\u591a\u6837\u4e14\u9ad8\u8d28\u91cf\u7684\u5408\u6210\u6570\u636e\u65b9\u9762\u5c55\u793a\u4e86\u663e\u8457\u6f5c\u529b\u3002Persona Hub\u7684\u591a\u529f\u80fd\u6027\u548c\u53ef\u6269\u5c55\u6027\u53ef\u4ee5\u63a8\u52a8LLM\u7814\u7a76\u548c\u5f00\u53d1\u7684\u8fdb\u6b65\uff0c\u6709\u53ef\u80fd\u5728\u5408\u6210\u6570\u636e\u521b\u5efa\u9886\u57df\u5e26\u6765\u8303\u5f0f\u8f6c\u53d8\u3002<\/p>\n\n\n\n<p>\u8be5\u65b9\u6cd5\u5229\u7528\u865a\u62df\u8eab\u4efd\u5f15\u5bfcLLM\u521b\u5efa\u5408\u6210\u6570\u636e\u3002\u901a\u8fc7\u5c06\u865a\u62df\u8eab\u4efd\u5d4c\u5165\u6570\u636e\u5408\u6210\u63d0\u793a\u4e2d\uff0cLLM\u53ef\u4ee5\u751f\u6210\u53cd\u6620\u8fd9\u4e9b\u865a\u62df\u8eab\u4efd\u4e2d\u5305\u542b\u7684\u77e5\u8bc6\u548c\u89c6\u89d2\u7684\u6570\u636e\u3002<\/p>\n\n\n\n<h4 class=\"wp-block-heading\">\u4e09\u3001\u6f5c\u5728\u5f71\u54cd<\/h4>\n\n\n\n<p>\u865a\u62df\u8eab\u4efd\u9a71\u52a8\u7684\u6570\u636e\u5408\u6210\u65b9\u6cd5\u4e3a\u751f\u6210\u591a\u6837\u5316\u7684\u5408\u6210\u6570\u636e\u63d0\u4f9b\u4e86\u53ef\u6269\u5c55\u7684\u89e3\u51b3\u65b9\u6848\u3002\u8fd9\u53ef\u4ee5\u663e\u8457\u589e\u5f3aLLM\u7684\u8bad\u7ec3\u3001\u6307\u4ee4\u8c03\u6574\u548c\u5728\u5404\u9886\u57df\u7684\u5e94\u7528\u3002\u7136\u800c\uff0c\u9700\u8981\u89e3\u51b3\u4f26\u7406\u95ee\u9898\u548c\u6f5c\u5728\u7684\u6ee5\u7528\uff0c\u4ee5\u786e\u4fdd\u8d1f\u8d23\u4efb\u7684\u5e94\u7528\u3002<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p>P.S., \u865a\u62df\u8eab\u4efd\u96c6\u5408\uff08Persona Hub\uff09\uff1a<a href=\"https:\/\/github.com\/tencent-ailab\/persona-hub\" target=\"_blank\" rel=\"noreferrer noopener\">https:\/\/github.com\/tencent-ailab\/persona-hub<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u817e\u8bafAI\u5b9e\u9a8c\u5ba4\u8fd1\u65e5\u53d1\u8868\u8bba\u6587\u300a\u5229\u752810\u4ebf\u4e2a\u865a\u62df\u8eab\u4efd\u6269\u5c55\u5408\u6210\u6570\u636e\u521b\u5efa\u300b\uff08Scaling Synthetic Dat [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[8],"tags":[39,86],"class_list":["post-4010","post","type-post","status-publish","format-standard","hentry","category-tech","tag-ai","tag-llm"],"rttpg_featured_image_url":null,"rttpg_author":{"display_name":"NullThought","author_link":"https:\/\/nullthought.net\/?author=1"},"rttpg_comment":0,"rttpg_category":"<a href=\"https:\/\/nullthought.net\/?cat=8\" rel=\"category\">Tech<\/a>","rttpg_excerpt":"\u817e\u8bafAI\u5b9e\u9a8c\u5ba4\u8fd1\u65e5\u53d1\u8868\u8bba\u6587\u300a\u5229\u752810\u4ebf\u4e2a\u865a\u62df\u8eab\u4efd\u6269\u5c55\u5408\u6210\u6570\u636e\u521b\u5efa\u300b\uff08Scaling Synthetic Dat&hellip;","_links":{"self":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts\/4010","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=4010"}],"version-history":[{"count":4,"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts\/4010\/revisions"}],"predecessor-version":[{"id":4018,"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts\/4010\/revisions\/4018"}],"wp:attachment":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4010"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4010"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4010"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}