{"id":117,"date":"2025-12-10T19:57:20","date_gmt":"2025-12-10T19:57:20","guid":{"rendered":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/?page_id=117"},"modified":"2025-12-10T20:07:53","modified_gmt":"2025-12-10T20:07:53","slug":"overview","status":"publish","type":"page","link":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/","title":{"rendered":"Overview"},"content":{"rendered":"\n<p>Modern video diffusion and world models can generate stunning, high-resolution videos\u2014but they are often either not physically grounded enough or too slow and expensive to use at scale. This project explores both sides of that problem across two semesters:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>2025 Spring: Richer physical representations through joint generation in a world model (Cosmos).<\/li>\n\n\n\n<li>2025 Fall: Efficient few-step generation in an open-source video model (Wan).<\/li>\n<\/ul>\n\n\n\n<p>In the first semester, we collaborated with NVIDIA on Cosmos, a world foundation model designed for training physical AI in simulation. We explored extending Cosmos beyond pure RGB and towards physical modalities such as depth and motion.<\/p>\n\n\n\n<p>In the second semester, in collaboration with NVIDIA and Pika, we shifted to making video diffusion more practical and deployable by improving sampling efficiency. We turned to Wan, a powerful open-source video diffusion model, and applied Distribution Matching Distillation (DMD) to turn Wan into a few-step video generator.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Modern video diffusion and world models can generate stunning, high-resolution videos\u2014but they are often either not physically grounded enough or too slow and expensive to use at scale. This project explores both sides of that problem across two semesters: In the first semester, we collaborated with NVIDIA on Cosmos, a world foundation model designed for &hellip; <\/p>\n<p class=\"link-more\"><a href=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Overview&#8221;<\/span><\/a><\/p>\n","protected":false},"author":257,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-117","page","type-page","status-publish","hentry"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Overview - Making Video Foundation Models Practical: From Physical Modalities to Fast Inference<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Overview - Making Video Foundation Models Practical: From Physical Modalities to Fast Inference\" \/>\n<meta property=\"og:description\" content=\"Modern video diffusion and world models can generate stunning, high-resolution videos\u2014but they are often either not physically grounded enough or too slow and expensive to use at scale. This project explores both sides of that problem across two semesters: In the first semester, we collaborated with NVIDIA on Cosmos, a world foundation model designed for &hellip; Continue reading &quot;Overview&quot;\" \/>\n<meta property=\"og:url\" content=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/\" \/>\n<meta property=\"og:site_name\" content=\"Making Video Foundation Models Practical: From Physical Modalities to Fast Inference\" \/>\n<meta property=\"article:modified_time\" content=\"2025-12-10T20:07:53+00:00\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"1 minute\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/\",\"url\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/\",\"name\":\"Overview - Making Video Foundation Models Practical: From Physical Modalities to Fast Inference\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/#website\"},\"datePublished\":\"2025-12-10T19:57:20+00:00\",\"dateModified\":\"2025-12-10T20:07:53+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Overview\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/#website\",\"url\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/\",\"name\":\"Making Video Foundation Models Practical: From Physical Modalities to Fast Inference\",\"description\":\"Authors: Sheldon Liang, Yinghao Zhang; Advisor: John Galeotti, Ethan He, Linnan Wang\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team2-2\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Overview - Making Video Foundation Models Practical: From Physical Modalities to Fast Inference","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/","og_locale":"en_US","og_type":"article","og_title":"Overview - Making Video Foundation Models Practical: From Physical Modalities to Fast Inference","og_description":"Modern video diffusion and world models can generate stunning, high-resolution videos\u2014but they are often either not physically grounded enough or too slow and expensive to use at scale. This project explores both sides of that problem across two semesters: In the first semester, we collaborated with NVIDIA on Cosmos, a world foundation model designed for &hellip; Continue reading \"Overview\"","og_url":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/","og_site_name":"Making Video Foundation Models Practical: From Physical Modalities to Fast Inference","article_modified_time":"2025-12-10T20:07:53+00:00","twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"1 minute"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/","url":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/","name":"Overview - Making Video Foundation Models Practical: From Physical Modalities to Fast Inference","isPartOf":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/#website"},"datePublished":"2025-12-10T19:57:20+00:00","dateModified":"2025-12-10T20:07:53+00:00","breadcrumb":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/"},{"@type":"ListItem","position":2,"name":"Overview"}]},{"@type":"WebSite","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/#website","url":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/","name":"Making Video Foundation Models Practical: From Physical Modalities to Fast Inference","description":"Authors: Sheldon Liang, Yinghao Zhang; Advisor: John Galeotti, Ethan He, Linnan Wang","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"}]}},"_links":{"self":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/wp-json\/wp\/v2\/pages\/117","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/wp-json\/wp\/v2\/users\/257"}],"replies":[{"embeddable":true,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/wp-json\/wp\/v2\/comments?post=117"}],"version-history":[{"count":3,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/wp-json\/wp\/v2\/pages\/117\/revisions"}],"predecessor-version":[{"id":126,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/wp-json\/wp\/v2\/pages\/117\/revisions\/126"}],"wp:attachment":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team2-2\/wp-json\/wp\/v2\/media?parent=117"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}