{"id":6719,"date":"2025-05-15T05:09:00","date_gmt":"2025-05-14T20:09:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6719"},"modified":"2025-05-10T17:11:13","modified_gmt":"2025-05-10T08:11:13","slug":"echoink-r1-exploring-audio-visual-reasoning-in-multimodal-llms-via-reinforcement-learning","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6719","title":{"rendered":"EchoInk-R1: Exploring Audio-Visual Reasoning in Multimodal LLMs via Reinforcement Learning"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>EchoInk-R1: Exploring Audio-Visual Reasoning in Multimodal LLMs via Reinforcement Learning\u00a0<\/strong>[108.7]<br>MLLM(Multimodal large language model)\u306f\u3001\u30c6\u30ad\u30b9\u30c8\u3001\u8996\u899a\u3001\u97f3\u58f0\u306b\u307e\u305f\u304c\u308b\u9ad8\u5ea6\u306a\u8a8d\u8b58\u3092\u6301\u3064\u304c\u3001\u69cb\u9020\u5316\u3055\u308c\u305f\u30af\u30ed\u30b9\u30e2\u30fc\u30c0\u30eb\u63a8\u8ad6\u306b\u82e6\u616e\u3059\u308b\u3002 MLLM\u306b\u304a\u3051\u308b\u305d\u306e\u3088\u3046\u306a\u63a8\u8ad6\u3092\u5f37\u5316\u3059\u308b\u5f37\u5316\u5b66\u7fd2\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3067\u3042\u308bEchoInk-R1\u3092\u7d39\u4ecb\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2505.04623v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2505.04623v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 07 May 2025 17:59:49 GMT)<\/li>\n\n\n\n<li>\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u306aReasoning\u30e2\u30c7\u30eb\u69cb\u7bc9\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u306e\u63d0\u6848\u3002\u300cwe adopt the Group Relative Policy Optimiza- tion (GRPO) reinforcement learning framework to the task of audio-image multiple-choice question answering in mul- timodal large language models (MLLMs)\u300d<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/HarryHsing\/EchoInk\">GitHub &#8211; HarryHsing\/EchoInk: EchoInk-R1: Exploring Audio-Visual Reasoning in Multimodal LLMs via Reinforcement Learning [\ud83d\udd25The Exploration of R1 for General Audio-Visual Reasoning with Qwen2.5-Omni]<\/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":[232,251],"class_list":["post-6719","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-lrm","tag-mllm"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6719","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=6719"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6719\/revisions"}],"predecessor-version":[{"id":6720,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6719\/revisions\/6720"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6719"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6719"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6719"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}