{"id":4126,"date":"2023-12-05T05:59:00","date_gmt":"2023-12-04T20:59:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4126"},"modified":"2023-12-05T05:59:00","modified_gmt":"2023-12-04T20:59:00","slug":"alignbench%e3%81%a8critiquellm","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4126","title":{"rendered":"AlignBench\u3068CRITIQUELLM"},"content":{"rendered":"\n<p>\u4e2d\u56fd\u8a9e\u306e\u30a2\u30e9\u30a4\u30e1\u30f3\u30c8\u8a55\u4fa1\u306e\u305f\u3081\u306e\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3068\u8a55\u4fa1\u30e2\u30c7\u30eb\u306e\u63d0\u6848\u3001\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/THUDM\/AlignBench\">GitHub &#8211; THUDM\/AlignBench: \u591a\u7ef4\u5ea6\u4e2d\u6587\u5bf9\u9f50\u8bc4\u6d4b\u57fa\u51c6 | Benchmarking Chinese Alignment of LLMs<\/a>\u3068<a href=\"https:\/\/github.com\/thu-coai\/CritiqueLLM\">GitHub &#8211; thu-coai\/CritiqueLLM<\/a><\/p>\n\n\n\n<p>\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306e\u898f\u6a21\u306f1000\u4ee5\u4e0b\u3068\u305d\u3053\u307e\u3067\u5927\u898f\u6a21\u3067\u306f\u306a\u3044\u304c\u3053\u306e\u624b\u306e\u57fa\u76e4\u3065\u304f\u308a\u306f\u65e5\u672c\u8a9e\u3067\u3082\u3084\u3063\u3066\u3044\u304d\u305f\u3044\u3068\u3053\u308d\u3002\u300cAdditionally, a systematic evaluation of 17 Chinese-supported LLMs was conducted to identify their levels of alignment.\u300d\u3068\u3042\u308b\u304c\u3001\u8a55\u4fa1\u7d50\u679c\u306fGPT-3.5\u3092\u8d85\u3048\u3066\u3044\u308b\u3082\u306e\u306f\u3042\u308b\u304c\u50c5\u5dee\u3067GPT-4\u306b\u306f\u53ca\u3093\u3067\u3044\u306a\u3044\u3001\u3068\u3044\u3046\u72b6\u6cc1\u306e\u3088\u3046\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>AlignBench: Benchmarking Chinese Alignment of Large Language Models&nbsp;<\/strong>[100.3]<br>\u4e2d\u56fd\u8a9e\u5927\u8a00\u8a9e\u30e2\u30c7\u30eb\u306e\u30a2\u30e9\u30a4\u30e1\u30f3\u30c8\u3092\u8a55\u4fa1\u3059\u308b\u305f\u3081\u306e\u7dcf\u5408\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u3042\u308bAlignBench\u3092\u7d39\u4ecb\u3059\u308b\u3002 \u7b46\u8005\u3089\u306e\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u306f,\u591a\u6b21\u5143LCM-as-Judge\u3068Chain-of-Thought\u3092\u7528\u3044\u3066,\u8aac\u660e\u3068\u6700\u7d42\u8a55\u4fa1\u3092\u8a55\u4fa1\u3068\u3057\u3066\u7528\u3044\u305f\u3002 \u307e\u305f, GPT-4\u306e\u8a55\u4fa1\u80fd\u529b\u306e95%\u3092\u56de\u5fa9\u3059\u308b\u5c02\u7528\u30b3\u30f3\u30d1\u30cb\u30aa\u30f3\u8a55\u4fa1\u5668\u3067\u3042\u308bCritiqueLLM\u3092\u958b\u767a\u3057\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2311.18743v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2311.18743v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Thu, 30 Nov 2023 17:41:30 GMT)<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>CritiqueLLM: Scaling LLM-as-Critic for Effective and Explainable Evaluation of Large Language Model Generation&nbsp;<\/strong>[89.8]<br>\u6211\u3005\u306f\u3001CrytiqueLLM\u3068\u547c\u3070\u308c\u308b\u65b0\u3057\u3044\u6279\u8a55\u751f\u6210\u30e2\u30c7\u30eb\u3092\u63d0\u6848\u3059\u308b\u3002 \u5b9f\u9a13\u7d50\u679c\u304b\u3089,GPT-4\u306b\u5339\u6575\u3059\u308b\u8a55\u4fa1\u6027\u80fd\u304c\u5f97\u3089\u308c\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2311.18702v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2311.18702v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Thu, 30 Nov 2023 16:52:42 GMT)<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>\u4e2d\u56fd\u8a9e\u306e\u30a2\u30e9\u30a4\u30e1\u30f3\u30c8\u8a55\u4fa1\u306e\u305f\u3081\u306e\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3068\u8a55\u4fa1\u30e2\u30c7\u30eb\u306e\u63d0\u6848\u3001\u30ea\u30dd\u30b8\u30c8\u30ea\u306fGitHub &#8211; THUDM\/AlignBench: \u591a\u7ef4\u5ea6\u4e2d\u6587\u5bf9\u9f50\u8bc4\u6d4b\u57fa\u51c6 | Benchmarking Chinese Alig &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=4126\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;AlignBench\u3068CRITIQUELLM&#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":[22,84,223,517],"class_list":["post-4126","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-alignment","tag-critic","tag-llm","tag-517"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4126","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=4126"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4126\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4126"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4126"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4126"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}