{"id":1076,"date":"2021-11-22T06:00:00","date_gmt":"2021-11-21T21:00:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=1076"},"modified":"2021-11-22T06:00:00","modified_gmt":"2021-11-21T21:00:00","slug":"raven-%e3%83%a2%e3%83%87%e3%83%ab%e3%81%af%e6%96%b0%e3%81%9f%e3%81%ab%e6%96%87%e6%9b%b8%e3%82%92%e4%bd%9c%e3%81%a3%e3%81%a6%e3%81%84%e3%82%8b%e3%81%ae%e3%81%8b%e3%80%81%e9%81%8e%e5%8e%bb%e3%81%ae","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=1076","title":{"rendered":"RAVEN: \u30e2\u30c7\u30eb\u306f\u65b0\u305f\u306b\u6587\u66f8\u3092\u4f5c\u3063\u3066\u3044\u308b\u306e\u304b\u3001\u5b66\u7fd2\u30c7\u30fc\u30bf\u3092\u30b3\u30d4\u30fc\u3057\u3066\u3044\u308b\u306e\u304b"},"content":{"rendered":"\n<ul class=\"wp-block-list\"><li><strong>How much do language models copy from their training data? Evaluating linguistic novelty in text generation using RAVEN&nbsp;<\/strong>[63.8]<br>\u73fe\u5728\u306e\u8a00\u8a9e\u30e2\u30c7\u30eb\u306f\u9ad8\u54c1\u8cea\u306a\u30c6\u30ad\u30b9\u30c8\u3092\u751f\u6210\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u308b\u3002 \u5f7c\u3089\u306f\u3001\u3053\u308c\u307e\u3067\u898b\u305f\u30c6\u30ad\u30b9\u30c8\u3092\u5358\u306b\u30b3\u30d4\u30fc\u3057\u3066\u3044\u308b\u304b\u3001\u305d\u308c\u3068\u3082\u4e00\u822c\u5316\u53ef\u80fd\u306a\u8a00\u8a9e\u7684\u62bd\u8c61\u5316\u3092\u5b66\u3093\u3060\u306e\u304b? \u672c\u7a3f\u3067\u306f\u3001\u751f\u6210\u3057\u305f\u30c6\u30ad\u30b9\u30c8\u306e\u65b0\u898f\u6027\u3092\u8a55\u4fa1\u3059\u308b\u305f\u3081\u306e\u5206\u6790\u30b9\u30a4\u30fc\u30c8\u3067\u3042\u308bRAVEN\u3092\u7d39\u4ecb\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2111.09509v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2111.09509v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Thu, 18 Nov 2021 04:07:09 GMT)<ul><li>\u65b0\u898f\u306b\u51fa\u3066\u304d\u305fn-gram\u306b\u6ce8\u76ee\u3057\u30c6\u30ad\u30b9\u30c8\u306e\u65b0\u898f\u6027\u3092\u8a55\u4fa1\u3059\u308b\u624b\u6cd5RAVEN\u3092\u958b\u767a\u3002\u8a00\u8a9e\u30e2\u30c7\u30eb\u304c\u751f\u6210\u3057\u305f\u30c6\u30ad\u30b9\u30c8\u306f\u5b66\u7fd2\u30c7\u30fc\u30bf\u306e\u30b3\u30d4\u30fc\u3067\u306f\u306a\u3044\u304b\uff1f\u3068\u3044\u3046\u7591\u554f\u306f\u6614\u304b\u3089\u6301\u3063\u3066\u3044\u3066\u8208\u5473\u6df1\u3044\u5185\u5bb9\u3002\u5c40\u6240\u7684\u306a\u69cb\u9020\u3067\u306f\u65b0\u898f\u6027\u304c\u4f4e\u3081\u3001\u5168\u4f53\u7684\u306a\u69cb\u9020\u3067\u306f\u65b0\u898f\u6027\u304c\u9ad8\u3081\u3001GPT-2\u3092\u5bfe\u8c61\u3068\u3057\u305f\u89e3\u6790\u3067\u306f\u610f\u5473\u7684\u554f\u984c\u304c\u6563\u898b\u3055\u308c\u305f\u3068\u306e\u3053\u3068\u3002<ul><li>\u975e\u5e38\u306b\u9577\u3044\u6587\u3092\u8907\u88fd\u3059\u308b\uff08\u4f8b\u5916\u7684\u306a\uff09\u4e8b\u8c61\u304c\u307f\u3089\u308c\u305f\u3068\u3042\u308a\u3001\u3053\u306e\u5370\u8c61\u304c\u30b3\u30d4\u30fc\u3092\u884c\u3063\u3066\u3044\u308b\u7591\u5ff5\u306b\u3064\u306a\u304c\u3063\u3066\u3044\u308b\u306e\u3067\u306f\u306a\u3044\u304b\u3068\u601d\u3046\u3002<\/li><\/ul><ul><li>RAVEN\u306e\u547d\u540d\u306f\u30a8\u30c9\u30ac\u30fc\u30fb\u30a2\u30e9\u30f3\u30fb\u30dd\u30fc\u306e<a href=\"https:\/\/ja.wikipedia.org\/wiki\/%E5%A4%A7%E9%B4%89\">\u5927\u9d09 &#8211; Wikipedia<\/a>\u304b\u3089\u3060\u305d\u3046\u3002<\/li><\/ul><\/li><li>\u30b3\u30fc\u30c9\u7b49\u306f\u516c\u958b\u4e88\u5b9a\u3068\u306e\u3053\u3068\u3002<\/li><\/ul><\/li><\/ul>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>How much do language models copy from their training data? Evaluating linguistic novelty in text generation us &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=1076\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;RAVEN: \u30e2\u30c7\u30eb\u306f\u65b0\u305f\u306b\u6587\u66f8\u3092\u4f5c\u3063\u3066\u3044\u308b\u306e\u304b\u3001\u5b66\u7fd2\u30c7\u30fc\u30bf\u3092\u30b3\u30d4\u30fc\u3057\u3066\u3044\u308b\u306e\u304b&#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":[170,567,634],"class_list":["post-1076","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-gigantic-language-model","tag-567","tag-634"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/1076","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=1076"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/1076\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1076"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1076"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1076"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}