{"id":9425,"date":"2026-06-21T02:19:39","date_gmt":"2026-06-20T18:19:39","guid":{"rendered":"https:\/\/infernews.com\/blog\/do-world-action-models-really-need-video-generation-or-just-image-editing\/"},"modified":"2026-06-21T02:53:42","modified_gmt":"2026-06-20T18:53:42","slug":"do-world-action-models-really-need-video-generation-or-just-image-editing","status":"publish","type":"post","link":"https:\/\/infernews.com\/blog\/do-world-action-models-really-need-video-generation-or-just-image-editing\/","title":{"rendered":"ImageWAM \u7528\u5716\u7247\u7de8\u8f2f\u505a\u6a5f\u68b0\u4eba\u6c7a\u7b56"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/06\/pasted-9faf9778e0cc.jpg\" alt=\"Repository image for yuyangalin\/ImageWAM\"\/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">ImageWAM \u662f\u4e00\u500b<strong>\u6a21\u578b\u8a13\u7df4\u8207\u8a55\u6e2c\u9805\u76ee<\/strong>\uff0c\u6838\u5fc3\u76ee\u6a19\u662f\u7528 image-editing foundation models \u53d6\u4ee3\u50b3\u7d71 World Action Models (WAMs) \u5e38\u898b\u7684\u5f71\u7247\u751f\u6210\u6d41\u7a0b\uff0c\u8655\u7406\u6a5f\u68b0\u4eba\u52d5\u4f5c\u9810\u6e2c\u53c8\u6162\u53c8\u91cd\u7684\u554f\u984c\u3002\u5b83\u7684\u5224\u65b7\u5f88\u9bae\u660e\uff1a\u8207\u5176\u751f\u6210\u4e00\u6bb5\u672a\u4f86\u756b\u9762\uff0c\u4e0d\u5982\u76f4\u63a5\u5f9e\u300c\u7576\u524d\u5f71\u50cf + \u6307\u4ee4\u300d\u62bd\u53d6\u8db3\u5920\u7684\u52d5\u4f5c\u7dda\u7d22\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u9019\u9805\u76ee\u628a\u5716\u7247\u7de8\u8f2f\u6a21\u578b\u7684\u4e2d\u9593\u8868\u5fb5\u62ff\u4f86\u505a robot action prediction\u3002\u6839\u64da\u9805\u76ee\u9801\u8cc7\u6599\uff0cImageWAM \u63a8\u8ad6\u6642\u4e0d\u4e00\u5b9a\u8981\u89e3\u78bc\u51fa\u7de8\u8f2f\u5f8c\u5f71\u50cf\uff0c\u800c\u662f\u4f7f\u7528\u55ae\u6b21 image editing forward step \u7522\u751f\u7684 KV caches\uff0c\u518d\u4ea4\u7d66 action expert \u751f\u6210\u672a\u4f86\u52d5\u4f5c\uff0c\u65b9\u5411\u4e0a\u6bd4\u591a\u5e40\u5f71\u7247\u9810\u6e2c\u66f4\u8f15\u91cf\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u5148\u770b FLUX.2 ImageWAM\uff0c\u56e0\u70ba\u5009\u5eab\u5df2\u8868\u660e\u5b83\u662f\u4e3b\u529b\u7248\u672c\uff0c\u4e26\u63d0\u4f9b 4B \u8207 9B \u8b8a\u9ad4\u3002\u4e4b\u5f8c\u518d\u6309\u624b\u4e0a\u8cc7\u6599\u8207\u7b97\u529b\uff0c\u6e96\u5099\u672c\u5730 datasets\u3001pretrained weights\u3001ActionDiT \u521d\u59cb\u5316\u6b0a\u91cd\uff0c\u7136\u5f8c\u5728 LIBERO\u3001LIBERO-plus \u6216 RoboTwin \u9019\u5e7e\u500b\u57fa\u6e96\u74b0\u5883\u505a\u8a13\u7df4\u8207\u8a55\u6e2c\u3002<\/p>\n\n\n<div\n    \tclass=\"align wp-block-vpb-video\"    id='vpbpVideoPlayer-1'\n    data-attributes='{&quot;source&quot;:&quot;https:\\\/\\\/zhangwenyao1.github.io\\\/ImageWAM\\\/assets\\\/ImageWAM_web.mp4&quot;,&quot;repeat&quot;:true,&quot;autoplay&quot;:true,&quot;muted&quot;:true,&quot;align&quot;:&quot;&quot;,&quot;poster&quot;:&quot;&quot;,&quot;controls&quot;:{&quot;play-large&quot;:true,&quot;restart&quot;:false,&quot;rewind&quot;:true,&quot;play&quot;:true,&quot;fast-forward&quot;:true,&quot;progress&quot;:true,&quot;current-time&quot;:true,&quot;duration&quot;:false,&quot;mute&quot;:true,&quot;volume&quot;:true,&quot;pip&quot;:false,&quot;airplay&quot;:false,&quot;settings&quot;:true,&quot;download&quot;:false,&quot;fullscreen&quot;:true},&quot;width&quot;:&quot;100%&quot;,&quot;radius&quot;:&quot;0px&quot;,&quot;resetOnEnd&quot;:false,&quot;autoHideControl&quot;:true,&quot;isSetup&quot;:false}'\n><\/div>\n\n\n<p class=\"wp-block-paragraph\">\u9019\u500b\u65b9\u5411\u4e0d\u53ea\u662f\u6982\u5ff5\u5be6\u9a57\u3002\u9805\u76ee\u9801\u5217\u51fa RoboTwin 2.0 \u70ba 93.38%\u3001LIBERO \u70ba 98.4%\u3001LIBERO-Plus \u70ba 83.1%\uff0c\u4e26\u63d0\u5230\u53ef\u7bc0\u7701 4.1\u00d7 FLOPs\u3001\u63a8\u8ad6\u5ef6\u9072\u52a0\u901f 84.7%\u3002\u9019\u4e9b\u6578\u5b57\u5f88\u5438\u5f15\uff0c\u4f46\u59cb\u7d42\u4ee5\u4f5c\u8005\u516c\u958b\u7684\u5be6\u9a57\u8a2d\u5b9a\u70ba\u6e96\uff0c\u82e5\u63db\u6210\u4e0d\u540c\u6a5f\u68b0\u4eba\u5e73\u53f0\u6216\u8cc7\u6599\u5206\u5e03\uff0c\u8868\u73fe\u4ecd\u8981\u518d\u9a57\u8b49\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u652f\u63f4\u591a\u500b\u76f8\u95dc\u6a21\u578b\uff1a<strong>FLUX.2 ImageWAM\u3001OmniGen2 ImageWAM\u3001Ovis-U1 ImageWAM<\/strong><\/li>\n\n\n\n<li>FLUX.2 \u63d0\u4f9b <strong>4B<\/strong> \u8207 <strong>9B<\/strong> \u7248\u672c\uff0cOvis-U1 \u8d70\u8f03\u7d30\u6a21\u578b\u8def\u7dda<\/li>\n\n\n\n<li>\u9069\u5408\u6a5f\u68b0\u4eba\u63a7\u5236\u3001world modeling\u3001action prediction \u7814\u7a76\u8207\u57fa\u6e96\u6e2c\u8a66<\/li>\n\n\n\n<li>\u91cd\u9ede\u4e0d\u662f\u751f\u6210\u6f02\u4eae\u756b\u9762\uff0c\u800c\u662f\u62bd\u53d6\u5c0d\u52d5\u4f5c\u6c7a\u7b56\u6709\u7528\u7684\u8b8a\u5316\u8cc7\u8a0a<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">\u6574\u9ad4\u4f86\u770b\uff0cImageWAM \u4e0d\u7b97\u9762\u5411\u4e00\u822c\u7528\u5bb6\u7684 AI \u5de5\u5177\uff0c\u66f4\u50cf\u7d66\u7814\u7a76\u8005\u8207\u5de5\u7a0b\u5718\u968a\u9a57\u8b49\u65b0\u8def\u7dda\u7684\u958b\u6e90\u9805\u76ee\u3002\u82e5\u4f60\u95dc\u5fc3 world action models \u662f\u5426\u4e00\u5b9a\u8981\u9760\u5f71\u7247\u751f\u6210\uff0c\u9019\u500b\u9805\u76ee\u63d0\u4f9b\u4e86\u4e00\u500b\u76f8\u7576\u5177\u9ad4\uff0c\u800c\u4e14\u6709\u57fa\u6e96\u6210\u7e3e\u652f\u6301\u7684\u53cd\u4f8b\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>GitHub\uff1a<\/strong> <a href=\"https:\/\/github.com\/yuyangalin\/ImageWAM\" rel=\"noopener noreferrer\" target=\"_blank\">https:\/\/github.com\/yuyangalin\/ImageWAM<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>\u9805\u76ee\uff1a<\/strong> <a href=\"https:\/\/zhangwenyao1.github.io\/ImageWAM\/\" rel=\"noopener noreferrer\" target=\"_blank\">https:\/\/zhangwenyao1.github.io\/ImageWAM\/<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>ImageWAM\u628a\u4e16\u754c\u52d5\u4f5c\u6a21\u578b\u6539\u6210\u5716\u7247\u7de8\u8f2f\u601d\u8def\uff0c\u5617\u8a66\u907f\u958b\u5f71\u7247\u751f\u6210\u7684\u9ad8\u6210\u672c\u3002\u5c0d\u6a5f\u68b0\u4eba\u63a7\u5236\u7814\u7a76\u8005\u4f86\u8aaa\uff0c\u9019\u500b\u65b9\u5411\u5e7e\u6709\u555f\u767c\u6027\u3002<\/p>\n","protected":false},"author":8,"featured_media":9424,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ai_generated_summary":"","footnotes":""},"categories":[133,31,30,38,150,157,120,76,127,149,184,186,199],"tags":[],"class_list":["post-9425","post","type-post","status-publish","format-standard","hentry","category-133","category-video","category-image","category-38","category-150","category-157","category-120","category-76","category-127","category-149","category-robotic","category-186","category-dataset-"],"_links":{"self":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/9425","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/comments?post=9425"}],"version-history":[{"count":3,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/9425\/revisions"}],"predecessor-version":[{"id":9431,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/9425\/revisions\/9431"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media\/9424"}],"wp:attachment":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media?parent=9425"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/categories?post=9425"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/tags?post=9425"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}