{"id":4206,"date":"2023-12-25T04:42:00","date_gmt":"2023-12-24T19:42:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4206"},"modified":"2023-12-25T04:42:00","modified_gmt":"2023-12-24T19:42:00","slug":"rest-meets-react","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4206","title":{"rendered":"ReST meets ReAct"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>ReST meets ReAct: Self-Improvement for Multi-Step Reasoning LLM Agent\u00a0<\/strong>[50.5]<br>\u5916\u90e8\u77e5\u8b58\u306b\u57fa\u3065\u3044\u3066\u63a8\u8ad6\u3068\u884c\u52d5\u3092\u884c\u3046ReAct-style LLM\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092\u958b\u767a\u3057\u305f\u3002 \u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092ReST\u30e9\u30a4\u30af\u306a\u624b\u6cd5\u3067\u6539\u826f\u3057,\u5f93\u6765\u306e\u8ecc\u9053\u4e0a\u3067\u53cd\u5fa9\u7684\u306b\u8a13\u7df4\u3059\u308b\u3002 \u5f15\u304d\u8d77\u3053\u3055\u308c\u305f\u5927\u304d\u306a\u30e2\u30c7\u30eb\u304b\u3089\u59cb\u307e\u308a\u3001\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306e\u305f\u3063\u305f2\u30a4\u30c6\u30ec\u30fc\u30b7\u30e7\u30f3\u306e\u5f8c\u306b\u3001\u5fae\u8abf\u6574\u3055\u308c\u305f\u5c0f\u3055\u306a\u30e2\u30c7\u30eb\u3092\u751f\u6210\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2312.10003v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2312.10003v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Fri, 15 Dec 2023 18:20:15 GMT)<\/li>\n\n\n\n<li>\u00a0Reinforced Self-Training (ReST)\u00a0\u3092\u9069\u7528\u3057\u305fReAct-style LLM agent\u306e\u63d0\u6848\u3002ReAct\u7684\u306a\u52d5\u304f\u3067\u4f5c\u3063\u305ftrajectory\u306e\u3046\u3061\u826f\u3044\u3082\u306e\u3092\u4f7f\u3063\u3066full fine-tuning\u3068\u304b\u306a\u308a\u306e\u8a08\u7b97\u91cf\u304c\u5fc5\u8981\u305d\u3046\u306a\u624b\u6cd5\u3002\u00a0\u5c11\u306a\u3044\u56de\u6570\u306e\u30a4\u30c6\u30ec\u30fc\u30b7\u30e7\u30f3\u3067\u826f\u3044\u6027\u80fd\u3092\u51fa\u305b\u308b\u3068\u306e\u3053\u3068\u3002<\/li>\n\n\n\n<li>\u300cemploying growing-batch reinforcement learning with AI feedback for continuous self-improvement and self-distillation.\u300d\u3068\u3042\u308b\u304c\u3001\u81ea\u5206\u3067\u5b66\u3093\u3067\u3044\u3051\u308bAI\u304c\u3066\u304d\u3064\u3064\u3042\u308b\u3093\u3058\u3083\u306a\u3044\u304b\u3068\u3044\u3046\u6c17\u3082\u3059\u308b\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":[42,223],"class_list":["post-4206","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-autonomous-agent","tag-llm"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4206","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=4206"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4206\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4206"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4206"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4206"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}