{"id":5787,"date":"2024-11-26T05:20:00","date_gmt":"2024-11-25T20:20:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5787"},"modified":"2024-11-26T05:20:00","modified_gmt":"2024-11-25T20:20:00","slug":"technical-report-enhancing-llm-reasoning-with-reward-guided-tree-search","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5787","title":{"rendered":"Technical Report: Enhancing LLM Reasoning with Reward-guided Tree Search \/ LLaVA-CoT\uff08LLaVA-o1\uff09"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Technical Report: Enhancing LLM Reasoning with Reward-guided Tree Search&nbsp;<\/strong>[95.1]<br>o1\u306e\u3088\u3046\u306a\u63a8\u8ad6\u30a2\u30d7\u30ed\u30fc\u30c1\u306f\u56f0\u96e3\u3067\u3001\u7814\u7a76\u8005\u306f\u3053\u306e\u30aa\u30fc\u30d7\u30f3\u306a\u7814\u7a76\u9818\u57df\u3092\u524d\u9032\u3055\u305b\u3088\u3046\u3068\u3055\u307e\u3056\u307e\u306a\u8a66\u307f\u3092\u884c\u3063\u3066\u304d\u305f\u3002 \u672c\u7a3f\u3067\u306f,\u5831\u916c\u8a98\u5c0e\u6728\u63a2\u7d22\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u3092\u7528\u3044\u3066,LLM\u306e\u63a8\u8ad6\u80fd\u529b\u3092\u9ad8\u3081\u308b\u305f\u3081\u306e\u4e88\u5099\u7684\u306a\u691c\u8a0e\u3092\u884c\u3046\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2411.11694v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2411.11694v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Mon, 18 Nov 2024 16:15:17 GMT)<\/li>\n\n\n\n<li>o1-like reasoning systems\u3092\u5b9f\u73fe\u3059\u308b\u305f\u3081\u306e\u691c\u8a0e\u3001\u300cIn this paper, we present a preliminary exploration into enhancing the reasoning abilities of LLMs through reward-guided tree search algorithms.\u300d\u3068\u306e\u3053\u3068\u3002Marco-o1\u306e\u5831\u544a\u3001DeepSeek-R1\u306e\u4e3b\u5f35<a href=\"https:\/\/techcrunch.com\/2024\/11\/20\/a-chinese-lab-has-released-a-model-to-rival-openais-o1\/?guccounter=1\">\uff08A Chinese lab has released a &#8216;reasoning&#8217; AI model to rival OpenAI&#8217;s o1 | TechCrunch\uff09<\/a>\u3092\u542b\u3081\u3001\u901f\u653b\u3067\u8fd1\u3044\u3082\u306e\u306e\u63d0\u6848\u304c\u59cb\u307e\u308b\u6fc0\u3057\u3044\u7af6\u4e89\u74b0\u5883\u3002\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u3067\u306e\u6709\u52b9\u6027\u3082\u5831\u544a\uff08\u4e0b\u8a18\uff09\u3055\u308c\u3066\u3044\u3066\u4eca\u5f8c\u304c\u697d\u3057\u307f\u3002<\/li>\n\n\n\n<li>\u30b5\u30fc\u30d9\u30a4\u306b\u8fd1\u3044\u304b\u3068\u601d\u3044\u304d\u3084\u5b9f\u9a13\u7d50\u679c\u306a\u3069\u3082\u3042\u308a\u53c2\u8003\u306b\u306a\u308b\u3002<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>LLaVA-o1: Let Vision Language Models Reason Step-by-Step\u00a0<\/strong>[33.7]<br>LLaVA-o1\u306f\u3001\u81ea\u5f8b\u7684\u306a\u591a\u6bb5\u968e\u63a8\u8ad6\u3092\u5b9f\u73fe\u3059\u308b\u305f\u3081\u306b\u8a2d\u8a08\u3055\u308c\u305f\u65b0\u3057\u3044VLM\u3067\u3042\u308b\u3002 \u30c1\u30a7\u30fc\u30f3\u30fb\u30aa\u30d6\u30fb\u30b7\u30fc\u30af\u30ec\u30c3\u30c8\u306e\u30d7\u30ed\u30f3\u30d7\u30c8\u3068\u306f\u7570\u306a\u308a\u3001LLaVA-o1\u306f\u72ec\u7acb\u306b\u8981\u7d04\u3001\u8996\u899a\u7684\u89e3\u91c8\u3001\u8ad6\u7406\u7684\u63a8\u8ad6\u3001\u7d50\u8ad6\u751f\u6210\u306e\u9010\u6b21\u7684\u306a\u6bb5\u968e\u306b\u95a2\u4e0e\u3059\u308b\u3002 100k\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30b5\u30f3\u30d7\u30eb\u3068\u5358\u7d14\u306a\u63a8\u8ad6\u6642\u9593\u30b9\u30b1\u30fc\u30ea\u30f3\u30b0\u6cd5\u306b\u3088\u308a\u3001LLaVA-o1\u306f\u30d9\u30fc\u30b9\u30e2\u30c7\u30eb\u3088\u308a\u30828.9%\u6027\u80fd\u304c\u5411\u4e0a\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2411.10440v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2411.10440v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Fri, 15 Nov 2024 18:58:31 GMT)<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/PKU-YuanGroup\/LLaVA-CoT\">GitHub &#8211; PKU-YuanGroup\/LLaVA-CoT: LLaVA-CoT, a visual language model capable of spontaneous, systematic reasoning<\/a>\u3001\u300cBased on recent feedback from social media platforms like X, we have decided to rename LLaVA-o1 to LLaVA-CoT.\u300d\u3068\u306e\u3053\u3068\u3002<\/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":[232,283],"class_list":["post-5787","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-lrm","tag-o1"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5787","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=5787"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5787\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5787"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5787"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5787"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}