{"id":7124,"date":"2025-07-25T05:23:00","date_gmt":"2025-07-24T20:23:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=7124"},"modified":"2025-07-20T14:28:59","modified_gmt":"2025-07-20T05:28:59","slug":"exaone-4-0-unified-large-language-models-integrating-non-reasoning-and-reasoning-modes","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=7124","title":{"rendered":"EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modes"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modes\u00a0<\/strong>[42.3]<br>EXAONE 4.0\u306f\u3001EXAONE 3.5\u306e\u512a\u308c\u305f\u4f7f\u3044\u3084\u3059\u3055\u3068EXAONE Deep\u306e\u9ad8\u5ea6\u306a\u63a8\u8ad6\u80fd\u529b\u306e\u4e21\u65b9\u3092\u9054\u6210\u3059\u308b\u305f\u3081\u306b\u3001\u975e\u63a8\u8ad6\u30e2\u30fc\u30c9\u3068\u63a8\u8ad6\u30e2\u30fc\u30c9\u3092\u7d71\u5408\u3057\u3066\u3044\u308b\u3002 EXAONE 4.0\u30b7\u30ea\u30fc\u30ba\u306f\u3001\u9ad8\u6027\u80fd\u306b\u6700\u9069\u5316\u3055\u308c\u305f\u4e2d\u578b32B\u30e2\u30c7\u30eb\u3068\u3001\u30aa\u30f3\u30c7\u30d0\u30a4\u30b9\u30a2\u30d7\u30ea\u30b1\u30fc\u30b7\u30e7\u30f3\u7528\u306b\u8a2d\u8a08\u3055\u308c\u305f\u5c0f\u578b1.2B\u30e2\u30c7\u30eb\u3067\u3042\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2507.11407v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2507.11407v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Tue, 15 Jul 2025 15:24:51 GMT)<\/li>\n\n\n\n<li>LLM\/LRM\u30cf\u30a4\u30d6\u30ea\u30c3\u30c9\u306aLG\u306e\u30e2\u30c7\u30eb\u3002\u300cUnified Mode Training In the combined dataset, the NON-REASONING data primarily consists of diverse tasks, while the REASONING data is centered on Math and Code domains. Rather than fine-tuning the two modes sequentially, we combine both modes and train them together.\u300d\u3068\u306e\u3053\u3068\u3002\u69cb\u7bc9\u904e\u7a0b\u306e\u300cAfter unified NON-REASONING\/REASONING mode fine-tuning, to address domain imbalance, we perform a second round of training using high-quality REASONING data from the Code and Tool Use domains, reusing these samples to further enhance the performance.\u300d\u304c\u8208\u5473\u6df1\u3044\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/huggingface.co\/LGAI-EXAONE\">LGAI-EXAONE (LG AI Research)<\/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":[223,232],"class_list":["post-7124","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-lrm"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7124","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=7124"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7124\/revisions"}],"predecessor-version":[{"id":7125,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7124\/revisions\/7125"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7124"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7124"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7124"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}