{"id":4967,"date":"2024-06-11T04:55:00","date_gmt":"2024-06-10T19:55:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4967"},"modified":"2024-06-11T04:55:00","modified_gmt":"2024-06-10T19:55:00","slug":"alice-in-wonderland-simple-tasks-showing-complete-reasoning-breakdown-in-state-of-the-art-large-language-models","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4967","title":{"rendered":"Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Models"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Models\u00a0<\/strong>[13.5]<br>\u5229\u7528\u53ef\u80fd\u306a\u6700\u5927\u898f\u6a21\u3067\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3055\u308c\u305f\u6700\u5148\u7aef\u30e2\u30c7\u30eb\u306e\u6a5f\u80fd\u3068\u63a8\u8ad6\u80fd\u529b\u306e\u5287\u7684\u306a\u7834\u58ca\u3092\u5b9f\u6f14\u3059\u308b\u3002 \u30e2\u30c7\u30eb\u306f\u9593\u9055\u3063\u305f\u89e3\u306b\u5f37\u3044\u81ea\u4fe1\u3092\u8868\u73fe\u3057\u3001\u3057\u3070\u3057\u3070\u975e\u611f\u899a\u7684\u306a\u300c\u63a8\u8ad6\u300d\u306e\u3088\u3046\u306a\u8aac\u660e\u3092\u63d0\u4f9b\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2406.02061v2\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2406.02061v2\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 05 Jun 2024 23:23:54 GMT)<\/li>\n\n\n\n<li>\u5f37\u529b\u306a\u306f\u305a\u306eLLM\u304c\u5358\u7d14\u306a\u554f\u984c\u300cAlice has N brothers and she also has M sisters. How many sisters does Alice\u2019s brother have?\u300d\u306b\u56de\u7b54\u3067\u304d\u306a\u3044\u3068\u3044\u3046\u6307\u6458\u3002MMLU\u306e\u7d50\u679c\u3068\u306e\u4e56\u96e2\u304c\u5927\u304d\u3044\u3002\n<ul class=\"wp-block-list\">\n<li>Leak\u3092\u542b\u3081\u8272\u3005\u306a\u554f\u984c\u304c\u3042\u308b\u3093\u3060\u308d\u3046\u3068\u601d\u3046\u3051\u3069\u3001\u300cWe also noticed during early experimentation that depending on choice of N and M and also the ordering of brothers and sisters in the sentence, the rate of correct responses may vary substantially.\u300d\u306f\u9762\u767d\u3044\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/LAION-AI\/AIW\">GitHub &#8211; LAION-AI\/AIW: Alice in Wonderland code base for experiments and raw experiments data<\/a><\/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":[223,517],"class_list":["post-4967","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-517"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4967","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=4967"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4967\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4967"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4967"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4967"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}