{"id":6309,"date":"2025-02-26T04:16:00","date_gmt":"2025-02-25T19:16:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6309"},"modified":"2025-02-26T04:16:00","modified_gmt":"2025-02-25T19:16:00","slug":"multilingual-machine-translation-with-open-large-language-models-at-practical-scale-an-empirical-study","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6309","title":{"rendered":"Multilingual Machine Translation with Open Large Language Models at Practical Scale: An Empirical Study\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Multilingual Machine Translation with Open Large Language Models at Practical Scale: An Empirical Study\u00a0<\/strong>[13.4]<br>GemmaX2-28\u306f\u300128\u8a00\u8a9e\u3067\u6700\u4e0a\u4f4d\u306e\u591a\u8a00\u8a9e\u7ffb\u8a33\u6027\u80fd\u3092\u9054\u6210\u3059\u308b9B\u30e2\u30c7\u30eb\u3067\u3042\u308b\u3002 GemmaX2-28 \u306f TowerInstruct \u3084 XALMA \u306a\u3069\u306e\u6700\u5148\u7aef (SOTA) \u30e2\u30c7\u30eb\u3088\u308a\u4e00\u8cab\u3057\u3066\u512a\u308c\u3066\u3044\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2502.02481v3\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2502.02481v3\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Fri, 07 Feb 2025 06:59:27 GMT)<\/li>\n\n\n\n<li>\u300cParallel-First Monolingual-Second (PFMS) data mixing strategy\u300d\u3092\u7528\u3044\u300cTo the best of our knowledge, GemmaX2-28-9B is the open model with the highest translation quality.\u300d\u3092\u4e3b\u5f35\u3059\u308b\u6a5f\u68b0\u7ffb\u8a33\u30e2\u30c7\u30eb\u306e\u63d0\u6848\u3002\u30c7\u30fc\u30bf\u306e\u30ec\u30b7\u30d4\u306b\u3088\u3063\u3066\u7ffb\u8a33\u6027\u80fd\u304c\u304b\u306a\u308a\u5909\u308f\u308b\u306e\u304c\u3068\u3066\u3082\u53c2\u8003\u306b\u306a\u308b\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/huggingface.co\/collections\/ModelSpace\/gemmax2-673714f5049bfa3a90bee6b6\">GemmaX2 &#8211; a ModelSpace Collection<\/a><\/li>\n<\/ul>\n\n\n\n<p><\/p>\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":[602],"class_list":["post-6309","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-602"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6309","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=6309"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6309\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6309"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6309"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6309"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}