{"id":7753,"date":"2025-11-11T06:17:00","date_gmt":"2025-11-10T21:17:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=7753"},"modified":"2025-11-09T09:25:07","modified_gmt":"2025-11-09T00:25:07","slug":"thinking-with-video-v-thinker","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=7753","title":{"rendered":"Thinking with Video, V-Thinker"},"content":{"rendered":"\n<p>\u63a8\u8ad6\u6642\u306b\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u306a\u30c7\u30fc\u30bf\u3092\u6d3b\u7528\u3059\u308b\u7814\u7a76\u304c\u9032\u3093\u3067\u3044\u308b\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm\u00a0<\/strong>[73.5]<br>\u30b7\u30f3\u30ad\u30f3\u30b0\u30fb\u30a6\u30a3\u30ba\u30fb\u30d3\u30c7\u30aa\u300d\u30d1\u30e9\u30c0\u30a4\u30e0\u306f\u3001\u8996\u899a\u7684\u30fb\u30c6\u30ad\u30b9\u30c8\u7684\u63a8\u8ad6\u3092\u7d71\u5408\u7684\u6642\u9593\u7684\u67a0\u7d44\u307f\u3067\u6a4b\u6e21\u3057\u3059\u308b\u3002 Sora-2\u306f\u30d3\u30b8\u30e7\u30f3\u4e2d\u5fc3\u306e\u30bf\u30b9\u30af\u306e\u6709\u80fd\u306a\u63a8\u8ad6\u8005\u3068\u3057\u3066\u78ba\u7acb\u3055\u308c\u3066\u3044\u308b\u3002 \u30c6\u30ad\u30b9\u30c8\u4e2d\u5fc3\u306e\u30bf\u30b9\u30af\u3067\u306f\u3001Sora-2\u306fMATH\u306792%\u3001MMMU\u306775.53%\u306e\u7cbe\u5ea6\u3092\u9054\u6210\u3057\u3066\u3044\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2511.04570v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2511.04570v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 06 Nov 2025 17:25:23 GMT)<\/li>\n\n\n\n<li>\u300cMoving beyond the traditional paradigms of \u201cThinking with Text\u201d (e g , Chain-of-Thought [3, 37]) and \u201cThinking with Images\u201d, we propose \u201cThinking with Video\u201d. It naturally enables human-like dynamic reasoning through video generation, such as drawing and imagination.\u300d\u3068\u52d5\u753b\u3092\u4f7f\u3063\u305f\u601d\u8003\u3002<\/li>\n\n\n\n<li>\u30d7\u30ed\u30b8\u30a7\u30af\u30c8\u30b5\u30a4\u30c8\u306f<a href=\"https:\/\/thinking-with-video.github.io\/\">Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm<\/a>\u3001\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/tongjingqi\/Thinking-with-Video\">GitHub &#8211; tongjingqi\/Thinking-with-Video: We introduce &#8220;Thinking with Video&#8221;, a new paradigm leveraging video generation for unified multimodal reasoning. Our VideoThinkBench shows that Sora-2 surpasses GPT5 by 10% on eyeballing puzzles and reaches 75% accuracy on MMMU, positioning video generation as a promising multimodal reasoning paradigm.<\/a><\/li>\n<\/ul>\n\n\n\n<p><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>V-Thinker: Interactive Thinking with Images\u00a0<\/strong>[22.6]<br>\u8996\u899a\u4e2d\u5fc3\u63a8\u8ad6\u306e\u6700\u8fd1\u306e\u9032\u6b69\u306f\u3001\u5927\u578b\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u30e2\u30c7\u30eb(LMM)\u306e\u305f\u3081\u306e\u6709\u671b\u306a\u300c\u30b7\u30f3\u30ad\u30f3\u30b0\u30fb\u30a6\u30a3\u30ba\u30fb\u30a4\u30e1\u30fc\u30b8\u300d\u30d1\u30e9\u30c0\u30a4\u30e0\u3092\u63a2\u6c42\u3057\u3066\u3044\u308b \u30a8\u30f3\u30c9\u30fb\u30c4\u30fc\u30fb\u30a8\u30f3\u30c9\u306e\u5f37\u5316\u5b66\u7fd2\u3092\u901a\u3058\u3066\u30a4\u30f3\u30bf\u30e9\u30af\u30c6\u30a3\u30d6\u306a\u8996\u899a\u4e2d\u5fc3\u306e\u601d\u8003\u3092\u53ef\u80fd\u306b\u3059\u308b\u6c4e\u7528\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u63a8\u8ad6\u30a2\u30b7\u30b9\u30bf\u30f3\u30c8\u3067\u3042\u308bV-Thinker\u3092\u63d0\u6848\u3059\u308b\u3002 V-Thinker\u306f\u3001\u4e00\u822c\u7684\u306a\u63a8\u8ad6\u30b7\u30ca\u30ea\u30aa\u3068\u5bfe\u8a71\u7684\u306a\u63a8\u8ad6\u30b7\u30ca\u30ea\u30aa\u306e\u4e21\u65b9\u306b\u304a\u3044\u3066\u3001\u5f37\u529b\u306aLMM\u30d9\u30fc\u30b9\u306e\u30d9\u30fc\u30b9\u30e9\u30a4\u30f3\u3092\u4e00\u8cab\u3057\u3066\u4e0a\u56de\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2511.04460v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2511.04460v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 06 Nov 2025 15:32:29 GMT)<\/li>\n\n\n\n<li>\u300cwe introduce V-Thinker, a general-purpose multimodal reasoning assistant that fosters interactive vision-centric thinking via end-to-end reinforcement training.\u300d\u3068\u8996\u899a\u3092\u6d3b\u7528\u3057\u305f\u601d\u8003\u3092\u884c\u3046\u30a2\u30b7\u30b9\u30bf\u30f3\u30c8\u306e\u63d0\u6848\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/We-Math\/V-Thinker\">GitHub &#8211; We-Math\/V-Thinker<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>\u63a8\u8ad6\u6642\u306b\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u306a\u30c7\u30fc\u30bf\u3092\u6d3b\u7528\u3059\u308b\u7814\u7a76\u304c\u9032\u3093\u3067\u3044\u308b\u3002<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[59,708],"class_list":["post-7753","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-chain-of-thought","tag-visual-chain-of-thought"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7753","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=7753"}],"version-history":[{"count":1,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7753\/revisions"}],"predecessor-version":[{"id":7754,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/7753\/revisions\/7754"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7753"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7753"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7753"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}