{"id":4997,"date":"2024-06-19T05:18:00","date_gmt":"2024-06-18T20:18:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4997"},"modified":"2024-06-19T05:18:00","modified_gmt":"2024-06-18T20:18:00","slug":"chain-of-preference-optimization-improving-chain-of-thought-reasoning-in-llms","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4997","title":{"rendered":"Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs\u00a0<\/strong>[37.1]<br>Tree-of- Thought (ToT) \u6cd5\u3067\u306f\u3001\u30c4\u30ea\u30fc\u63a2\u7d22\u3092\u7528\u3044\u3066\u63a8\u8ad6\u7a7a\u9593\u3092\u5e83\u7bc4\u56f2\u306b\u63a2\u7d22\u3057\u3001CoT\u30c7\u30b3\u30fc\u30c7\u30a3\u30f3\u30b0\u304c\u898b\u843d\u3068\u3057\u3066\u3057\u307e\u3046\u304b\u3082\u3057\u308c\u306a\u3044\u63a8\u8ad6\u7d4c\u8def\u3092\u3088\u308a\u3088\u304f\u898b\u3064\u3051\u308b\u3002 ToT\u3067\u69cb\u7bc9\u3055\u308c\u305f\u691c\u7d22\u30c4\u30ea\u30fc\u3092\u5229\u7528\u3057\u305f\u7d30\u8abf\u6574\u8a00\u8a9e\u30e2\u30c7\u30eb(LLMs)\u306b\u3088\u308a\u3001CoT\u306f\u540c\u69d8\u306e\u3042\u308b\u3044\u306f\u3088\u308a\u826f\u3044\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u3092\u5b9f\u73fe\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u308b\u3002 \u3053\u308c\u306fCPO(Chain of Preference Optimization)\u306b\u3088\u3063\u3066\u5b9f\u73fe\u3055\u308c\u3001LLM\u306fCoT\u63a8\u8ad6\u30d1\u30b9\u306e\u5404\u30b9\u30c6\u30c3\u30d7\u3092ToT\u306e\u30b9\u30c6\u30c3\u30d7\u3068\u6574\u5217\u3059\u308b\u3088\u3046\u306b\u5fae\u8abf\u6574\u3055\u308c\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2406.09136v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2406.09136v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 13 Jun 2024 14:07:02 GMT)<\/li>\n\n\n\n<li>ToT\u306e\u7d4c\u8def\u3092\u7528\u3044\u3066\u30c1\u30e5\u30fc\u30cb\u30f3\u30b0\uff08DPO\uff09\u3059\u308bChain of Preference Optimization\u306b\u3088\u308a\u3001CoT\u3092\u8d85\u3048\u3001ToT\u306b\u8fd1\u3044\u6027\u80fd\u3092\u9054\u6210\u3001\u8a08\u7b97\u6642\u9593\u3082\u6291\u3048\u3089\u308c\u308b\u3068\u3059\u308b\u8ad6\u6587\u3002\u30b7\u30f3\u30d7\u30eb\u306bSFT\u3059\u308bTS-SFT\u3088\u308a\u6027\u80fd\u304c\u9ad8\u3044\u306e\u304c\u610f\u5916\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/sail-sg\/CPO\">GitHub &#8211; sail-sg\/CPO<\/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":[60,223],"class_list":["post-4997","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-chain","tag-llm"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4997","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=4997"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4997\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4997"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4997"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4997"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}