{"id":7374,"date":"2025-09-02T06:08:00","date_gmt":"2025-09-01T21:08:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=7374"},"modified":"2025-09-01T14:19:39","modified_gmt":"2025-09-01T05:19:39","slug":"memento-fine-tuning-llm-agents-without-fine-tuning-llms","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=7374","title":{"rendered":"Memento: Fine-tuning LLM Agents without Fine-tuning LLMs\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Memento: Fine-tuning LLM Agents without Fine-tuning LLMs\u00a0<\/strong>[36.3]<br>\u672c\u7a3f\u3067\u306f,\u9069\u5fdc\u578b\u5927\u8a00\u8a9e\u30e2\u30c7\u30eb(LLM)\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u305f\u3081\u306e\u65b0\u3057\u3044\u5b66\u7fd2\u30d1\u30e9\u30c0\u30a4\u30e0\u3092\u63d0\u6848\u3059\u308b\u3002 \u672c\u624b\u6cd5\u306f,\u30e1\u30e2\u30ea\u30d9\u30fc\u30b9\u306e\u30aa\u30f3\u30e9\u30a4\u30f3\u5f37\u5316\u5b66\u7fd2\u306b\u3088\u308a,\u4f4e\u30b3\u30b9\u30c8\u3067\u9023\u7d9a\u7684\u306a\u9069\u5fdc\u3092\u53ef\u80fd\u306b\u3059\u308b\u3002 \u6211\u3005\u306f\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u30e2\u30c7\u30eb\u3092,GAIA\u691c\u8a3c\u3067\u30c8\u30c3\u30d71\u306b\u9054\u3059\u308bMemento\u3068\u3044\u3046\u30c7\u30a3\u30fc\u30d7\u30ea\u30b5\u30fc\u30c1\u74b0\u5883\u3067\u30a4\u30f3\u30b9\u30bf\u30f3\u30b9\u5316\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2508.16153v2\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2508.16153v2\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Mon, 25 Aug 2025 13:32:12 GMT)<\/li>\n\n\n\n<li>\u300cMemento formalises deep research agents as a memory-based Markov Decision Process (MDP) and implements it within a planner\u2013executor framework, leveraging an episodic case bank to record and retrieve trajectories for continual policy improvement.\u300d\u3068\u3044\u3046\u30e1\u30e2\u30ea\u6a5f\u69cb\u3092\u6301\u3064\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u306e\u63d0\u6848\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/Agent-on-the-Fly\/Memento\">GitHub &#8211; Agent-on-the-Fly\/Memento: Official Code of Memento: Fine-tuning LLM Agents without Fine-tuning LLMs<\/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":[686,223,244],"class_list":["post-7374","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-deepresearch","tag-llm","tag-memory"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7374","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=7374"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7374\/revisions"}],"predecessor-version":[{"id":7375,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7374\/revisions\/7375"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7374"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7374"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7374"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}