{"id":6513,"date":"2025-04-03T05:16:00","date_gmt":"2025-04-02T20:16:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6513"},"modified":"2025-04-03T05:16:00","modified_gmt":"2025-04-02T20:16:00","slug":"trade-offs-in-large-reasoning-models-an-empirical-analysis-of-deliberative-and-adaptive-reasoning-over-foundational-capabilities","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6513","title":{"rendered":"Trade-offs in Large Reasoning Models: An Empirical Analysis of Deliberative and Adaptive Reasoning over Foundational Capabilities"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Trade-offs in Large Reasoning Models: An Empirical Analysis of Deliberative and Adaptive Reasoning over Foundational Capabilities\u00a0<\/strong>[101.8]<br>\u8fd1\u5e74\u306eLRM(Large Reasoning Models)\u306e\u9032\u6b69\u306f\u3001\u7279\u6b8a\u63a8\u8ad6\u30bf\u30b9\u30af\u306b\u304a\u3044\u3066\u9855\u8457\u306a\u6027\u80fd\u3092\u793a\u3057\u3066\u3044\u308b\u3002 \u8b70\u8ad6\u7684\u63a8\u8ad6\u80fd\u529b\u306e\u7372\u5f97\u306f, LRM\u306e\u57fa\u790e\u7684\u80fd\u529b\u3092\u5927\u5e45\u306b\u4f4e\u4e0b\u3055\u305b\u308b\u3053\u3068\u3092\u793a\u3059\u3002 \u9069\u5fdc\u63a8\u8ad6(Zero-Thinking, Less-Thinking, Summary-Thinking)\u304c\u3053\u308c\u3089\u306e\u6b20\u70b9\u3092\u52b9\u679c\u7684\u306b\u8efd\u6e1b\u3067\u304d\u308b\u3053\u3068\u3092\u793a\u3057\u307e\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2503.17979v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2503.17979v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Sun, 23 Mar 2025 08:18:51 GMT)<\/li>\n\n\n\n<li>\u300cThe overall results of different LRMs under the Zero-Thinking, Summary-Thinking and Summary-Thinking-Plus mode for the evaluation of foundational capabilities.\u300d\u306e\u88685\u306e\u7d50\u679c\u304c\u975e\u5e38\u306b\u8208\u5473\u6df1\u3044\u3002\u63a8\u8ad6\u306b\u30d1\u30ef\u30fc\u3092\u304b\u3051\u308c\u3070\u3088\u3044\u3068\u3044\u3046\u308f\u3051\u3067\u3082\u306a\u304f\u9069\u5fdc\u578b\u6226\u7565\u306e\u91cd\u8981\u6027\u304c\u3088\u304f\u308f\u304b\u308b\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/SCIR-SC-Qiaoban-Team\/FreeEvalLM\">GitHub &#8211; SCIR-SC-Qiaoban-Team\/FreeEvalLM<\/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":[232,298],"class_list":["post-6513","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-lrm","tag-overthinking"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6513","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=6513"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6513\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6513"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6513"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6513"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}