{"id":5818,"date":"2024-12-03T05:04:00","date_gmt":"2024-12-02T20:04:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5818"},"modified":"2024-12-03T05:04:00","modified_gmt":"2024-12-02T20:04:00","slug":"enhancing-llm-reasoning-via-critique-models-with-test-time-and-training-time-supervision","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5818","title":{"rendered":"Enhancing LLM Reasoning via Critique Models with Test-Time and Training-Time Supervision\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Enhancing LLM Reasoning via Critique Models with Test-Time and Training-Time Supervision\u00a0<\/strong>[120.4]<br>\u672c\u7a3f\u3067\u306f\u3001\u63a8\u8ad6\u3068\u6279\u5224\u30e2\u30c7\u30eb\u306e\u5f79\u5272\u3092\u5206\u96e2\u3059\u308b2\u4eba\u30d7\u30ec\u30a4\u30e4\u30d1\u30e9\u30c0\u30a4\u30e0\u3092\u63d0\u6848\u3059\u308b\u3002 \u307e\u305a\u3001\u6279\u5224\u30c7\u30fc\u30bf\u3092\u53ce\u96c6\u3059\u308b\u81ea\u52d5\u5316\u304a\u3088\u3073\u30b9\u30b1\u30fc\u30e9\u30d6\u30eb\u306a\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3067\u3042\u308bAutoMathCritique\u3092\u63d0\u6848\u3059\u308b\u3002 \u30c6\u30b9\u30c8\u6642\u9593\u306b\u304a\u3051\u308b\u96e3\u89e3\u306a\u30af\u30a8\u30ea\u306b\u5bfe\u3059\u308b\u30a2\u30af\u30bf\u30fc\u306e\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u3092,\u6279\u5224\u30e2\u30c7\u30eb\u304c\u4e00\u8cab\u3057\u3066\u6539\u5584\u3059\u308b\u3053\u3068\u304c\u5b9f\u8a3c\u3055\u308c\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2411.16579v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2411.16579v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Mon, 25 Nov 2024 17:11:54 GMT)<\/li>\n\n\n\n<li>\u300cflawed reasoning path construction, critique generation, and data filtering\u300d\u306e3\u30b9\u30c6\u30fc\u30b8\u304b\u3089\u306a\u308b\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30afAutoMathCritique\u3067\u30c7\u30fc\u30bf\u3092\u69cb\u7bc9\u3001fine tuning\u3059\u308b\u3068\u3068\u3082\u306b\u3001\u300cMotivated by the insights of test-time, we introduce the critique model into the actor model\u2019s exploration and learning process, introducing a critique-in-the-loop self-improvement method\u300d\u3092\u9069\u7528\u3057\u3066\u52b9\u679c\u3092\u78ba\u8a8d\u3002\u00a0critique model\u306e\u6709\u52b9\u6027\u304c\u5206\u304b\u308b\u7d50\u679c\u306b\u898b\u3048\u308b\uff08\u304c\u3001\u3053\u306e\u69cb\u7bc9\u306f\u5bb9\u6613\u3067\u306f\u306a\u3044\u304b\u3082\u3057\u308c\u306a\u3044\uff09<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/mathcritique.github.io\/\">AutoMathCritique<\/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":[84,223,356,390],"class_list":["post-5818","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-critic","tag-llm","tag-self-x","tag-synthetic-data"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5818","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=5818"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5818\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5818"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5818"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5818"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}