{"id":4082,"date":"2023-11-23T04:57:00","date_gmt":"2023-11-22T19:57:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4082"},"modified":"2023-11-23T04:57:00","modified_gmt":"2023-11-22T19:57:00","slug":"instrusum","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4082","title":{"rendered":"INSTRUSUM\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Benchmarking Generation and Evaluation Capabilities of Large Language Models for Instruction Controllable Summarization\u00a0<\/strong>[136.2]<br>\u547d\u4ee4\u5236\u5fa1\u53ef\u80fd\u306a\u30c6\u30ad\u30b9\u30c8\u8981\u7d04\u306e\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb(LLM)\u3092\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3059\u308b\u3002 \u672c\u7814\u7a76\u306f,LLM\u306b\u304a\u3044\u3066,\u547d\u4ee4\u5236\u5fa1\u53ef\u80fd\u306a\u30c6\u30ad\u30b9\u30c8\u8981\u7d04\u304c\u4f9d\u7136\u3068\u3057\u3066\u56f0\u96e3\u306a\u8ab2\u984c\u3067\u3042\u308b\u3053\u3068\u3092\u793a\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2311.09184v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2311.09184v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 15 Nov 2023 18:25:26 GMT)<\/li>\n\n\n\n<li>\u5236\u5fa1\u3055\u308c\u305f\u30c6\u30ad\u30b9\u30c8\u8981\u7d04\u306e\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3002GPT-4\u3067\u3042\u308c\u3070\u53ef\u80fd\u306a\u306e\u304b\u3068\u601d\u3046\u3068\u3053\u308d\u3060\u304c\u300cWe found that several LLMs have already shown promising performance in generating ins-controllable summaries.\u300d\u3067\u3042\u308b\u3082\u306e\u306e\u300cHowever, they lack robust holistic capabilities for this task since they still make a considerable amount of errors in their summaries and they can not reliability evaluate the different candidate summaries for the same data example\u300d\u3068\u96e3\u3057\u3044\u3088\u3046\u3002\uff08\u3082\u3068\u304b\u3089\u7c21\u5358\u306a\u30bf\u30b9\u30af\u3067\u306f\u306a\u3044\u3067\u306f\u306a\u3044\u3082\u306e\u306e\uff09LLM\u3067\u3042\u308c\u3070\u5bfe\u5fdc\u53ef\u80fd\u3068\u8a00\u3044\u5207\u308c\u306a\u3044\u306e\u306f\u8208\u5473\u6df1\u3044\u7d50\u679c\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/yale-nlp\/InstruSum\">GitHub &#8211; yale-nlp\/InstruSum<\/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":[223,491,517,640],"class_list":["post-4082","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-491","tag-517","tag-640"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4082","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=4082"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4082\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4082"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4082"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4082"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}