{"categories":["CreativeVision"],"width":"100%","blog_title":"SB Intuitions TECH BLOG","provider_url":"https://hatena.blog","url":"https://www.sbintuitions.co.jp/blog/entry/2026/08/18/083500","author_name":"shiqi_yang_sbint","height":"190","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Fwww.sbintuitions.co.jp%2Fblog%2Fentry%2F2026%2F08%2F18%2F083500\" title=\"Few-Step World Action Models: Action-Side Distillation of Fast-WAM - SB Intuitions TECH BLOG\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","blog_url":"https://www.sbintuitions.co.jp/blog/","image_url":"https://media-cdn.atlassian.com/file/055ffbcf-a1ab-45f0-a760-a4d166f70c8a/image/cdn?allowAnimated=true&client=b3298b63-1b24-4944-a161-3272496a8d53&collection=contentId-1778450441&height=125&max-age=2592000&mode=full-fit&source=mediaCard&token=eyJhbGciOiJIUzI1NiJ9.eyJpc3MiOiJiMzI5OGI2My0xYjI0LTQ5NDQtYTE2MS0zMjcyNDk2YThkNTMiLCJhY2Nlc3MiOnsidXJuOmZpbGVzdG9yZTpjb2xsZWN0aW9uOmNvbnRlbnRJZC0xNzc4NDUwNDQxIjpbInJlYWQiXX0sImV4cCI6MTc4NjYxNDkyNSwibmJmIjoxNzg2NjEyMDQ1LCJhYUlkIjoiNzEyMDIwOjM4YTlmYWM4LWJhMTctNGNiYi1iNDZkLWE3YjQxOTdlZjM2MyIsImh0dHBzOi8vaWQuYXRsYXNzaWFuLmNvbS9hcHBBY2NyZWRpdGVkIjpmYWxzZSwiYXV0aFR5cGUiOiJzZXNzaW9uIn0.G8nUWOjbsi4Uj8md9eFsJQj5CFdgYWAf99cYyjeIxH8&width=760#media-blob-url=true&id=055ffbcf-a1ab-45f0-a760-a4d166f70c8a&clientId=b3298b63-1b24-4944-a161-3272496a8d53&contextId=contentId-1778450441&collection=contentId-1778450441","author_url":"https://blog.hatena.ne.jp/shiqi_yang_sbint/","title":"Few-Step World Action Models: Action-Side Distillation of Fast-WAM","version":"1.0","provider_name":"Hatena Blog","type":"rich","published":"2026-08-18 08:35:00","description":"(Japanese version follows English one/\u82f1\u8a9e\u7248\u306e\u5f8c\u306b\u65e5\u672c\u8a9e\u7248\u304c\u7d9a\u304d\u307e\u3059\u3002) Contributor Shiqi YangCreative Vision Team, Multimodal AI Department Motivation World Action Models (WAMs) have recently emerged as a promising direction for embodied intelligence [1, 2, 3]. They connect visual world models with robot action ge\u2026"}