{"id":4169,"date":"2023-12-19T05:47:00","date_gmt":"2023-12-18T20:47:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4169"},"modified":"2023-12-19T05:47:00","modified_gmt":"2023-12-18T20:47:00","slug":"beyond-human-data","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4169","title":{"rendered":"Beyond Human Data"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models\u00a0<\/strong>[105.5]<br>\u4eba\u70ba\u7684\u306a\u30c7\u30fc\u30bf\u306b\u57fa\u3065\u304f\u5fae\u8abf\u6574\u8a00\u8a9e\u30e2\u30c7\u30eb(LM)\u304c\u666e\u53ca\u3057\u3066\u3044\u308b\u3002 \u6211\u3005\u306f\u3001\u30b9\u30ab\u30e9\u30fc\u30d5\u30a3\u30fc\u30c9\u30d0\u30c3\u30af\u306b\u30a2\u30af\u30bb\u30b9\u53ef\u80fd\u306a\u30bf\u30b9\u30af\u306b\u304a\u3044\u3066\u3001\u4eba\u9593\u306e\u30c7\u30fc\u30bf\u3092\u8d85\u3048\u308b\u3053\u3068\u304c\u3067\u304d\u308b\u304b\u3069\u3046\u304b\u691c\u8a0e\u3059\u308b\u3002 ReST$EM$\u306f\u30e2\u30c7\u30eb\u30b5\u30a4\u30ba\u306b\u597d\u9069\u306b\u30b9\u30b1\u30fc\u30eb\u3057\u3001\u4eba\u9593\u306e\u30c7\u30fc\u30bf\u306e\u307f\u306b\u3088\u308b\u5fae\u8abf\u6574\u3092\u5927\u5e45\u306b\u4e0a\u56de\u3063\u3066\u3044\u308b\u3053\u3068\u304c\u308f\u304b\u3063\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2312.06585v2\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2312.06585v2\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Tue, 12 Dec 2023 23:16:16 GMT)<\/li>\n\n\n\n<li>LLM\u3078\u306e\u5408\u6210\u30c7\u30fc\u30bf\u9069\u7528\u304c\u6709\u52b9\u304b\u691c\u8a3c\u3057\u305f\u8ad6\u6587\u3002\u751f\u6210\u2192\u30d5\u30a3\u30eb\u30bf\u2192finetune\u2192\u751f\u6210\u2192\u30fb\u30fb\u30fb\u3068\u3044\u3046\u81ea\u5df1\u5b66\u7fd2\u5f62\u5f0f\u3002<\/li>\n\n\n\n<li>\u6570\u5b66\u3084\u30b3\u30fc\u30c9\u751f\u6210\u3067\u6709\u52b9\u306a\u306e\u306f\u305d\u3046\u306a\u306e\u304b\u306a\u3068\u601d\u3046\u3002limitation\u3068\u3057\u3066\u6319\u304c\u3063\u3066\u3044\u305f\u300cSecond, ReST\ud835\udc38\ud835\udc40 also requires access to a manually-designed or learned reward function, ideally one that can be computed automatically.\u300d\u306f\u91cd\u8981\u3002<\/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":[99,223,390],"class_list":["post-4169","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-data-centric-ai","tag-llm","tag-synthetic-data"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4169","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=4169"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4169\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4169"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4169"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4169"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}