{"id":6591,"date":"2025-04-24T04:43:00","date_gmt":"2025-04-23T19:43:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6591"},"modified":"2025-04-24T04:43:00","modified_gmt":"2025-04-23T19:43:00","slug":"mm-ifengine-towards-multimodal-instruction-following","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6591","title":{"rendered":"MM-IFEngine: Towards Multimodal Instruction Following\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>MM-IFEngine: Towards Multimodal Instruction Following\u00a0<\/strong>[85.9]<br>\u9ad8\u54c1\u8cea\u306a\u30a4\u30e1\u30fc\u30b8\u30a4\u30f3\u30b9\u30c8\u30e9\u30af\u30b7\u30e7\u30f3\u30da\u30a2\u3092\u751f\u6210\u3059\u308b\u30d1\u30a4\u30d7\u30e9\u30a4\u30f3\u3067\u3042\u308bMM-IFEngine\u3092\u63d0\u6848\u3059\u308b\u3002 MM-IFInstruct-23k\u306fSFT(Supervised Fine-Tuning)\u306b\u9069\u3057\u3066\u3044\u308b\u304c\u3001DPO(Direct Preference Optimization)\u306e\u305f\u3081\u306bMM-IFDPO-23k\u3068\u3057\u3066\u62e1\u5f35\u3055\u308c\u3066\u3044\u308b\u3002 \u307e\u305f\u3001MM-IFEval\u306f\u3001\u56f0\u96e3\u3067\u591a\u69d8\u306a\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u547d\u4ee4\u8ffd\u5f93\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u3042\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2504.07957v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2504.07957v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 10 Apr 2025 17:59:12 GMT)<\/li>\n\n\n\n<li>\u300cthe instruction-following ability of Multimodal Large Language Models\u300d\u306e\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3068\u30e2\u30c7\u30eb\uff08\u516c\u958b\u30e2\u30c7\u30eb\u30d9\u30fc\u30b9\uff09\u306e\u63d0\u6848\u3002\u5546\u7528\u30e2\u30c7\u30eb\u306e\u5f37\u529b\u3055\u304c\u76ee\u7acb\u3064\u3002\u307e\u305f\u3001\u300cDPO using MM-IFDPO-23k significantly surpasses SFT on MMIFInstruct-23k\u300d\u306f\u8208\u5473\u6df1\u3044\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/SYuan03\/MM-IFEngine\">GitHub &#8211; SYuan03\/MM-IFEngine: MM-IFEngine: Towards Multimodal Instruction Following<\/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":[517],"class_list":["post-6591","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-517"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6591","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=6591"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6591\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6591"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6591"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6591"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}