{"id":1956,"date":"2022-06-13T06:00:00","date_gmt":"2022-06-12T21:00:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=1956"},"modified":"2022-06-13T06:00:00","modified_gmt":"2022-06-12T21:00:00","slug":"beyond-the-imitation-game-benchmark-big-bench","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=1956","title":{"rendered":"Beyond the Imitation Game benchmark (BIG-bench)"},"content":{"rendered":"\n<ul class=\"wp-block-list\"><li><strong>Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models&nbsp;<\/strong>[645.0]<br>\u8a00\u8a9e\u30e2\u30c7\u30eb\u306f\u3001\u898f\u6a21\u304c\u5927\u304d\u304f\u306a\u308b\u306b\u3064\u308c\u3066\u91cf\u7684\u6539\u5584\u3068\u65b0\u3057\u3044\u8cea\u7684\u80fd\u529b\u306e\u4e21\u65b9\u3092\u793a\u3059\u3002 \u30d3\u30c3\u30b0\u30d9\u30f3\u30c1\u306f204\u306e\u30bf\u30b9\u30af\u3067\u69cb\u6210\u3055\u308c\u3001132\u306e\u6a5f\u95a2\u3067442\u4eba\u306e\u8457\u8005\u304c\u8ca2\u732e\u3057\u3066\u3044\u308b\u3002 \u6211\u3005\u306f,OpenAI\u306eGPT\u30e2\u30c7\u30eb,Google\u5185\u90e8\u306e\u9ad8\u5bc6\u5ea6\u30c8\u30e9\u30f3\u30b9\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3,BIG\u30d9\u30f3\u30c1\u4e0a\u306e\u30b9\u30a4\u30c3\u30c1\u578b\u30b9\u30d1\u30fc\u30b9\u30c8\u30e9\u30f3\u30b9\u306e\u6319\u52d5\u3092\u8a55\u4fa1\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2206.04615v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2206.04615v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Thu, 9 Jun 2022 17:05:34 GMT)<ul><li>204\u30bf\u30b9\u30af\u304b\u3089\u306a\u308b\u5927\u898f\u6a21\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3068\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb\u3092\u7528\u3044\u305f\u8a55\u4fa1\u306e\u5831\u544a\u3002<ul><li>\u8457\u8005\u304c\u6975\u3081\u3066\u591a\u304f\u3001\u30b9\u30b3\u30a2=645\u306ffugumt.com\u5185\u3067\u6700\u9ad8\u3002<\/li><\/ul><\/li><li>\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb\u3092\u7528\u3044\u305fAI\u3067\u3082\u7dcf\u5408\u7684\u306b\u306f\u4eba\u9593\u306f\u53ca\u3070\u306a\u3044\u3001\u958b\u767a\u8005\uff08OpenAI or Google\uff09\u304c\u7570\u306a\u3063\u3066\u3044\u3066\u3082AI\u306f\u4f3c\u305f\u6319\u52d5\u3092\u793a\u3059\u306a\u3069\u975e\u5e38\u306b\u8208\u5473\u6df1\u3044\u3002<ul><li>\u6700\u8fd1\u306e\u300c\u30d1\u30e9\u30e1\u30fc\u30bf\u6570\u306e\u5897\u52a0\uff08\u5927\u898f\u6a21\u5316\uff09\u306b\u3088\u308aAI\u306e\u6027\u80fd\u306f\u4eba\u9593\u306b\u4e26\u3079\u308b\u300d\u3068\u3044\u3046\u96f0\u56f2\u6c17\u3068\u306e\u6574\u5408\u6027\u306f\u8b0e\u3067\u30015\u30da\u30fc\u30b8\u306ePaLM\u306e\u7d50\u679c\u306b\u3088\u3063\u3066\u89e3\u91c8\u304c\u5909\u308f\u308b\u30ec\u30d9\u30eb\u306e\u5f71\u97ff\u304c\u3042\u308a\u305d\u3046\u3002PaLM\u304c\u7121\u3051\u308c\u3070\u7dcf\u5408\u7684\u306b\u4eba\u9593\u306e\u30ec\u30d9\u30eb\u306b\u5230\u9054\u3059\u308b\u9053\u306e\u308a\u306f\u9577\u305d\u3046\u3060\u304c\u3001PaLM\u306e\u7d50\u679c\u304c\u3042\u308b\u3068\u884c\u3051\u305d\u3046\u306b\u601d\u3048\u308b\u3002<\/li><li>\u300cDuring the writing of this work, results for the Pathways Language Model (PaLM) on BIG-bench were published (Chowdhery et al., 2022) and we included them in Figure 1. Most of the analysis in the paper is focused on results from models we evaluated BIG-bench on.\u300d\u3068\u306e\u3053\u3068\u3067\u3001\u3053\u306e\u5206\u91ce\u306f\u9032\u5c55\u304c\u901f\u3059\u304e\u308b\u3068\u3044\u3046\u5370\u8c61<\/li><\/ul><\/li><\/ul><ul><li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/google\/BIG-bench\">GitHub &#8211; google\/BIG-bench: Beyond the Imitation Game collaborative benchmark for measuring and extrapolating the capabilities of language models<\/a><\/li><\/ul><\/li><\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models&nbsp;[645.0]\u8a00\u8a9e\u30e2\u30c7\u30eb &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=1956\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;Beyond the Imitation Game benchmark (BIG-bench)&#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":[491,567],"class_list":["post-1956","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-491","tag-567"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/1956","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=1956"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/1956\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1956"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1956"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1956"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}