PageSourceSearch

https://pabloruizponce.com/papers

html pabloruizponce.com collected 2026-10-03 23:18:34 UTC 34,432 bytes, 2 lines download raw bytes

1<!DOCTYPE html><html lang="en"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="preload" href="/_next/static/media/e11418ac562b8ac1-s.p.woff2" as="font" crossorigin="" type="font/woff2"/><link rel="stylesheet" href="/_next/static/css/d9dcc5c2af16d07f.css?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" data-precedence="next"/><link rel="preload" as="script" fetchPriority="low" href="/_next/static/chunks/webpack-8226c7bf9e0898d5.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5"/>
1<script src="/_next/static/chunks/4bd1b696-409494caf8c83275.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" async=""></script>
1<script src="/_next/static/chunks/255-1600c87892b01378.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" async=""></script>
1<script src="/_next/static/chunks/main-app-f810cc9959894cd9.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" async=""></script>
1<script src="/_next/static/chunks/619-f072ac750404f9da.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" async=""></script>
1<script src="/_next/static/chunks/720-1feb7901ab6ca450.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" async=""></script>
1<script src="/_next/static/chunks/app/layout-fe7e933091b9aa41.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" async=""></script>
1<script src="/_next/static/chunks/356-983d6c86a9d42db5.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" async=""></script>
1<script src="/_next/static/chunks/app/papers/page-9bcdd6b969ebb432.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" async=""></script>
1<link rel="preload" href="/umami.js" as="script"/><meta name="next-size-adjust" content=""/><title>Pablo Ruiz-Ponce</title><meta name="description" content="Personal website of Pablo Ruiz-Ponce"/><link rel="canonical" href="https://pabloruizponce.com"/><meta property="og:title" content="Pablo Ruiz-Ponce"/><meta property="og:description" content="Personal website of Pablo Ruiz-Ponce"/><meta property="og:url" content="https://pabloruizponce.com"/><meta property="og:site_name" content="Pablo Ruiz-Ponce"/><meta property="og:locale" content="en_US"/><meta property="og:image:alt" content="Pablo Ruiz Ponce"/><meta property="og:image:type" content="image/png"/><meta property="og:image" content="https://pabloruizponce.com/opengraph-image?5ecce4c2dd6e2ff4"/><meta property="og:image:width" content="1200"/><meta property="og:image:height" content="630"/><meta property="og:type" content="website"/><meta name="twitter:card" content="summary_large_image"/><meta name="twitter:site" content="@pabloruizponce"/><meta name="twitter:creator" content="@pabloruizponce"/><meta name="twitter:title" content="Pablo Ruiz-Ponce"/><meta name="twitter:description" content="Personal website of Pablo Ruiz-Ponce"/><meta name="twitter:image" content="https://pabloruizponce.com/opengraph-image"/><meta name="twitter:image:alt" content="Pablo Ruiz-Ponce"/><meta name="twitter:image:width" content="1200"/><meta name="twitter:image:height" content="630"/><link rel="icon" href="/favicon.ico" type="image/x-icon" sizes="256x256"/><link rel="icon" href="/favicon.ico"/>
1<script src="/_next/static/chunks/polyfills-42372ed130431b0a.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" noModule=""></script>
1</head><body class="__className_3a0388"><div hidden=""><!--$--><!--/$--></div><div class="sticky top-0 flex flex-col items-center justify-centers bg-white shadow-md w-full z-50"><nav class="flex flex-col lg:flex-row lg:py-4 items-center justify-center transition-all duration-300 ease-in-out"><button class="lg:hidden group inline-flex w-12 h-12 text-slate-800 bg-white text-center items-center justify-center rounded transition" aria-pressed="false"><span class="sr-only">Menu</span><svg class="w-6 h-6 fill-current pointer-events-none" viewBox="0 0 16 16" xmlns="http://www.w3.org/2000/svg"><rect class="origin-center -translate-y-[5px] translate-x-[7px] transition-all duration-300 ease-[cubic-bezier(.5,.85,.25,1.1)] group-[[aria-pressed=true]]:translate-x-0 group-[[aria-pressed=true]]:translate-y-0 group-[[aria-pressed=true]]:rotate-[315deg]" y="7" width="9" height="2" rx="1"></rect><rect class="origin-center transition-all duration-300 ease-[cubic-bezier(.5,.85,.25,1.8)] group-[[aria-pressed=true]]:rotate-45" y="7" width="16" height="2" rx="1"></rect><rect class="origin-center translate-y-[5px] transition-all duration-300 ease-[cubic-bezier(.5,.85,.25,1.1)] group-[[aria-pressed=true]]:translate-y-0 group-[[aria-pressed=true]]:rotate-[135deg]" y="7" width="9" height="2" rx="1"></rect></svg></button><div class="flex flex-col mb-2 lg:mb-0 lg:flex-row gap-4 lg:gap-10 text-center transition-all duration-300 ease-in-out lg:opacity-100 lg:max-h-screen opacity-0 max-h-0 overflow-hidden"><a class=" text-lg" href="/">Home</a><a class="font-bold text-primary text-lg" href="/papers">Papers</a><a class=" text-lg" href="/contact">Contact</a><a target="_blank" rel="noopener noreferrer" class="text-xl" href="https://files.pabloruizponce.com/static/resume.pdf">CV</a></div></nav></div><div class="py-10 w-full flex flex-col items-center"><div class="container px-4 mb-4 lg:max-w-3xl justify-start items-start"><h1 class="text-4xl font-bold">Papers</h1><p>Here are some of the most recent and relevant papers I have published. To see the full list of papers I have contributed to, check my<!-- --> <a target="_blank" rel="noopener noreferrer" class="text-primary hover:underline" data-umami-event="papers-google-scholar-click" href="https://scholar.google.com/citations?user=yourID">Google Scholar</a>.</p></div><!--$--><section class="container flex flex-col gap-4 px-4 lg:max-w-3xl"><div class="bg-white rounded-xl shadow border border-gray-200 overflow-hidden undefined"><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-5-image-click" href="papers/Interact2Ar"><img alt="Graphical abstract of the paper" loading="lazy" width="3918" height="1897" decoding="async" data-nimg="1" class="w-full object-cover p-5" style="color:transparent" srcSet="/_next/image?url=https%3A%2F%2Ffiles.pabloruizponce.com%2Fpapers%2FInteract2Ar%2Fcover.webp&amp;w=3840&amp;q=75 1x" src="/_next/image?url=https%3A%2F%2Ffiles.pabloruizponce.com%2Fpapers%2FInteract2Ar%2Fcover.webp&amp;w=3840&amp;q=75"/> </a><div class="w-full"><div class="border-t border-gray-200"></div></div><div class="flex flex-col gap-4 py-6 px-8 justify-start items-start text-left"><div class="flex flex-col"><span class="text-lg text-gray-500">CVPR 2026</span><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-5-title-click" href="papers/Interact2Ar"><strong class="text-xl hover:text-primary transition-all duration-300">Interact2Ar: Full-Body Human-Human Interaction Generation via Autoregressive Diffusion Models
2</strong></a></div><p>We introduce Interact2Ar, the first text-conditioned autoregressive diffusion model for generating full-body human-human interactions. Unlike previous approaches that generate entire sequences simultaneously, Interact2Ar utilizes a novel memory strategy and dedicated parallel branches to synthesize high-fidelity motions with detailed hand kinematics. This enables adaptive capabilities, such as temporal composition and real-time adaptation to disturbances. We also introduce robust evaluators and extended metrics to assess full-body interaction quality.</p><div class="flex flex-wrap gap-4"><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-5-web-click" href="papers/Interact2Ar"><img alt="Web" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/web.svg"/></a><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-5-paper-click" href="https://arxiv.org/abs/2512.19692"><img alt="Paper" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/paper.svg"/>
2</a><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-5-github-click" href="https://github.com/pabloruizponce/Interact2Ar"><img alt="GitHub" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/github.svg"/></a></div></div></div><div class="bg-white rounded-xl shadow border border-gray-200 overflow-hidden undefined"><div class="flex flex-col gap-4 py-6 px-8 justify-start items-start text-left"><div class="flex flex-col"><span class="text-lg text-gray-500">CVPR 2025</span><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-4-title-click" href="papers/MixerMDM"><strong class="text-xl hover:text-primary transition-all duration-300">MixerMDM: Learnable Composition of Human Motion Diffusion Models</strong></a></div><p>We introduce MixerMDM, the first learnable approach for combining multiple pre-trained text-conditioned motion diffusion models. Unlike previous static methods, MixerMDM learns a dynamic mixing strategy to adaptively combine models based on their generated motions and input conditions. This enables fine-grained and consistent individual controllability over human-human interaction motion generation. We also propose a novel evaluation technique that measures mixing quality by assessing individual and interaction alignment.</p><div class="flex flex-wrap gap-4"><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-4-web-click" href="papers/MixerMDM"><img alt="Web" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/web.svg"/></a><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-4-paper-click" href="https://arxiv.org/abs/2504.01019"><img alt="Paper" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/paper.svg"/></a><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-4-github-click" href="https://github.com/pabloruizponce/MixerMDM"><img alt="GitHub" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/github.svg"/></a></div></div></div><div class="bg-white rounded-xl shadow border border-gray-200 overflow-hidden undefined"><div class="flex flex-col gap-4 py-6 px-8 justify-start items-start text-left"><div class="flex flex-col"><span class="text-lg text-gray-500">CVPR 2024 HuMoGen Workshop</span><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-1-title-click" href="https://pabloruizponce.github.io/in2IN/"><strong class="text-xl hover:text-primary transition-all duration-300">in2IN: Leveraging individual Information to Generate Human INteractions</strong></a></div><p>We present in2IN, a diffusion model for human-human motion generation that uses both overall and individual action descriptions to improve the realism and diversity of interactions. By augmenting the InterHuman dataset with individual textual annotations using a large language model, in2IN achieves state-of-the-art results. To further boost intra-person diversity, we propose DualMDM, a model composition method that combines motions from in2IN with a single-person model trained on HumanML3D. Together, they enhance control over individual behavior while preserving coherent interaction dynamics.</p><div class="flex flex-wrap gap-4"><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-1-web-click" href="https://pabloruizponce.github.io/in2IN/"><img alt="Web" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/web.svg"/></a><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-1-paper-click" href="https://arxiv.org/abs/2404.09988"><img alt="Paper" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/paper.svg"/>
2</a><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-1-github-click" href="https://github.com/pabloruizponce/in2IN"><img alt="GitHub" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/github.svg"/></a></div></div></div><div class="bg-white rounded-xl shadow border border-gray-200 overflow-hidden undefined"><div class="flex flex-col gap-4 py-6 px-8 justify-start items-start text-left"><div class="flex flex-col"><span class="text-lg text-gray-500">UPC Commons</span><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-2-title-click" href="https://upcommons.upc.edu/handle/2117/408075"><strong class="text-xl hover:text-primary transition-all duration-300">Text-driven multi-human motion generation</strong></a></div><p>In this thesis, we introduce a novel Diffusion Model incorporating a Transformer-based architecture. This model is conditioned using textual descriptions of both the motion interactions and the individual motions within these interactions. By focusing on the individual components of the interaction, our method achieves more precise conditioning in the generation of these specific motions. Concurrently, the textual descriptions of the overall interaction enable our model to effectively capture the interplay between individual motions.</p><div class="flex flex-wrap gap-4"><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-2-web-click" href="https://upcommons.upc.edu/handle/2117/408075"><img alt="Web" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/web.svg"/></a><a target="_blank" rel="noopener noreferrer" data-umami-event="paper-card-2-paper-click" href="https://upcommons.upc.edu/handle/2117/408075"><img alt="Paper" loading="lazy" width="8" height="8" decoding="async" data-nimg="1" class="w-8 h-8 hover:scale-110 transition-all duration-300" style="color:transparent" src="icons/paper.svg"/></a></div></div></div></section><!--/$--></div><!--$--><!--/$--><section aria-label="Notifications alt+T" tabindex="-1" aria-live="polite" aria-relevant="additions text" aria-atomic="false"></section>
2<script src="/_next/static/chunks/webpack-8226c7bf9e0898d5.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5" id="_R_" async=""></script>
2<script>(self.__next_f=self.__next_f||[]).push([0])</script>
2<script>self.__next_f.push([1,"1:\"$Sreact.fragment\"\n2:I[41402,[\"619\",\"static/chunks/619-f072ac750404f9da.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"720\",\"static/chunks/720-1feb7901ab6ca450.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"177\",\"static/chunks/app/layout-fe7e933091b9aa41.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\"],\"\"]\n3:I[248,[\"619\",\"static/chunks/619-f072ac750404f9da.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"720\",\"static/chunks/720-1feb7901ab6ca450.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"177\",\"static/chunks/app/layout-fe7e933091b9aa41.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\"],\"NavBar\"]\n4:I[9766,[],\"\"]\n5:I[98924,[],\"\"]\n6:I[25294,[\"619\",\"static/chunks/619-f072ac750404f9da.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"720\",\"static/chunks/720-1feb7901ab6ca450.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"177\",\"static/chunks/app/layout-fe7e933091b9aa41.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\"],\"Toaster\"]\n7:I[52619,[\"619\",\"static/chunks/619-f072ac750404f9da.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"356\",\"static/chunks/356-983d6c86a9d42db5.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"228\",\"static/chunks/app/papers/page-9bcdd6b969ebb432.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\"],\"\"]\n8:\"$Sreact.suspense\"\na:I[24431,[],\"OutletBoundary\"]\nc:I[15278,[],\"AsyncMetadataOutlet\"]\ne:I[24431,[],\"ViewportBoundary\"]\n10:I[24431,[],\"MetadataBoundary\"]\n12:I[57150,[],\"\"]\n:HL[\"/_next/static/media/e11418ac562b8ac1-s.p.woff2\",\"font\",{\"crossOrigin\":\"\",\"type\":\"font/woff2\"}]\n:HL[\"/_next/static/css/d9dcc5c2af16d07f.css?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"style\"]\n"])</script>
2<script>self.__next_f.push([1,"0:{\"P\":null,\"b\":\"fnUZRjdLOuLFZZp3fQFKq\",\"p\":\"\",\"c\":[\"\",\"papers\"],\"i\":false,\"f\":[[[\"\",{\"children\":[\"papers\",{\"children\":[\"__PAGE__\",{}]}]},\"$undefined\",\"$undefined\",true],[\"\",[\"$\",\"$1\",\"c\",{\"children\":[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/_next/static/css/d9dcc5c2af16d07f.css?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"precedence\":\"next\",\"crossOrigin\":\"$undefined\",\"nonce\":\"$undefined\"}]],[\"$\",\"html\",null,{\"lang\":\"en\",\"children\":[[\"$\",\"head\",null,{\"children\":[\"$\",\"$L2\",null,{\"defer\":true,\"src\":\"/umami.js\",\"data-website-id\":\"03269dfa-4a8e-4da3-8165-c9fe2cd935a6\"}]}],[\"$\",\"body\",null,{\"className\":\"__className_3a0388\",\"children\":[[\"$\",\"div\",null,{\"className\":\"sticky top-0 flex flex-col items-center justify-centers bg-white shadow-md w-full z-50\",\"children\":[\"$\",\"$L3\",null,{}]}],[\"$\",\"$L4\",null,{\"parallelRouterKey\":\"children\",\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L5\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":404}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],[]],\"forbidden\":\"$undefined\",\"unauthorized\":\"$undefined\"}],[\"$\",\"$L6\",null,{\"richColors\":true}]]}]]}]]}],{\"children\":[\"papers\",[\"$\",\"$1\",\"c\",{\"children\":[null,[\"$\",\"$L4\",null,{\"parallelRouterKey\":\"children\",\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L5\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":\"$undefined\",\"forbidden\":\"$undefined\",\"unauthorized\":\"$undefined\"}]]}],{\"children\":[\"__PAGE__\",[\"$\",\"$1\",\"c\",{\"children\":[[\"$\",\"div\",null,{\"className\":\"py-10 w-full flex flex-col items-center\",\"children\":[[\"$\",\"div\",null,{\"className\":\"container px-4 mb-4 lg:max-w-3xl justify-start items-start\",\"children\":[[\"$\",\"h1\",null,{\"className\":\"text-4xl font-bold\",\"children\":\"Papers\"}],[\"$\",\"p\",null,{\"children\":[\"Here are some of the most recent and relevant papers I have published. To see the full list of papers I have contributed to, check my\",\" \",[\"$\",\"$L7\",null,{\"href\":\"https://scholar.google.com/citations?user=yourID\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"className\":\"text-primary hover:underline\",\"data-umami-event\":\"papers-google-scholar-click\",\"children\":\"Google Scholar\"}],\".\"]}]]}],[\"$\",\"$8\",null,{\"fallback\":[\"$\",\"p\",null,{\"children\":\"Loading papers...\"}],\"children\":\"$L9\"}]]}],null,[\"$\",\"$La\",null,{\"children\":[\"$Lb\",[\"$\",\"$Lc\",null,{\"promise\":\"$@d\"}]]}]]}],{},null,false]},null,false]},null,false],[\"$\",\"$1\",\"h\",{\"children\":[null,[[\"$\",\"$Le\",null,{\"children\":\"$Lf\"}],[\"$\",\"meta\",null,{\"name\":\"next-size-adjust\",\"content\":\"\"}]],[\"$\",\"$L10\",null,{\"children\":[\"$\",\"div\",null,{\"hidden\":true,\"children\":[\"$\",\"$8\",null,{\"fallback\":null,\"children\":\"$L11\"}]}]}]]}],false]],\"m\":\"$undefined\",\"G\":[\"$12\",[]],\"s\":false,\"S\":true}\n"])</script>
2<script>self.__next_f.push([1,"f:[[\"$\",\"meta\",\"0\",{\"charSet\":\"utf-8\"}],[\"$\",\"meta\",\"1\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}]]\nb:null\n"])</script>
2<script>self.__next_f.push([1,"13:I[80622,[],\"IconMark\"]\n"])</script>
2<script>self.__next_f.push([1,"d:{\"metadata\":[[\"$\",\"title\",\"0\",{\"children\":\"Pablo Ruiz-Ponce\"}],[\"$\",\"meta\",\"1\",{\"name\":\"description\",\"content\":\"Personal website of Pablo Ruiz-Ponce\"}],[\"$\",\"link\",\"2\",{\"rel\":\"canonical\",\"href\":\"https://pabloruizponce.com\"}],[\"$\",\"meta\",\"3\",{\"property\":\"og:title\",\"content\":\"Pablo Ruiz-Ponce\"}],[\"$\",\"meta\",\"4\",{\"property\":\"og:description\",\"content\":\"Personal website of Pablo Ruiz-Ponce\"}],[\"$\",\"meta\",\"5\",{\"property\":\"og:url\",\"content\":\"https://pabloruizponce.com\"}],[\"$\",\"meta\",\"6\",{\"property\":\"og:site_name\",\"content\":\"Pablo Ruiz-Ponce\"}],[\"$\",\"meta\",\"7\",{\"property\":\"og:locale\",\"content\":\"en_US\"}],[\"$\",\"meta\",\"8\",{\"property\":\"og:image:alt\",\"content\":\"Pablo Ruiz Ponce\"}],[\"$\",\"meta\",\"9\",{\"property\":\"og:image:type\",\"content\":\"image/png\"}],[\"$\",\"meta\",\"10\",{\"property\":\"og:image\",\"content\":\"https://pabloruizponce.com/opengraph-image?5ecce4c2dd6e2ff4\"}],[\"$\",\"meta\",\"11\",{\"property\":\"og:image:width\",\"content\":\"1200\"}],[\"$\",\"meta\",\"12\",{\"property\":\"og:image:height\",\"content\":\"630\"}],[\"$\",\"meta\",\"13\",{\"property\":\"og:type\",\"content\":\"website\"}],[\"$\",\"meta\",\"14\",{\"name\":\"twitter:card\",\"content\":\"summary_large_image\"}],[\"$\",\"meta\",\"15\",{\"name\":\"twitter:site\",\"content\":\"@pabloruizponce\"}],[\"$\",\"meta\",\"16\",{\"name\":\"twitter:creator\",\"content\":\"@pabloruizponce\"}],[\"$\",\"meta\",\"17\",{\"name\":\"twitter:title\",\"content\":\"Pablo Ruiz-Ponce\"}],[\"$\",\"meta\",\"18\",{\"name\":\"twitter:description\",\"content\":\"Personal website of Pablo Ruiz-Ponce\"}],[\"$\",\"meta\",\"19\",{\"name\":\"twitter:image\",\"content\":\"https://pabloruizponce.com/opengraph-image\"}],[\"$\",\"meta\",\"20\",{\"name\":\"twitter:image:alt\",\"content\":\"Pablo Ruiz-Ponce\"}],[\"$\",\"meta\",\"21\",{\"name\":\"twitter:image:width\",\"content\":\"1200\"}],[\"$\",\"meta\",\"22\",{\"name\":\"twitter:image:height\",\"content\":\"630\"}],[\"$\",\"link\",\"23\",{\"rel\":\"icon\",\"href\":\"/favicon.ico\",\"type\":\"image/x-icon\",\"sizes\":\"256x256\"}],[\"$\",\"link\",\"24\",{\"rel\":\"icon\",\"href\":\"/favicon.ico\"}],[\"$\",\"$L13\",\"25\",{}]],\"error\":null,\"digest\":\"$undefined\"}\n"])</script>
2<script>self.__next_f.push([1,"11:\"$d:metadata\"\n"])</script>
2<script>self.__next_f.push([1,"14:I[81356,[\"619\",\"static/chunks/619-f072ac750404f9da.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"356\",\"static/chunks/356-983d6c86a9d42db5.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\",\"228\",\"static/chunks/app/papers/page-9bcdd6b969ebb432.js?dpl=dpl_8sMa7AwvfddCWcZsB4hzmdUMFbc5\"],\"Image\"]\n"])</script>
2<script>self.__next_f.push([1,"9:[\"$\",\"section\",null,{\"className\":\"container flex flex-col gap-4 px-4 lg:max-w-3xl\",\"children\":[[\"$\",\"div\",\"5\",{\"className\":\"bg-white rounded-xl shadow border border-gray-200 overflow-hidden undefined\",\"children\":[[[\"$\",\"$L7\",null,{\"href\":\"papers/Interact2Ar\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-5-image-click\",\"children\":[[\"$\",\"$L14\",null,{\"alt\":\"Graphical abstract of the paper\",\"src\":\"https://files.pabloruizponce.com/papers/Interact2Ar/cover.webp\",\"className\":\"w-full object-cover p-5\",\"width\":3918,\"height\":1897}],\" \"]}],[\"$\",\"div\",null,{\"className\":\"w-full\",\"children\":[\"$\",\"div\",null,{\"className\":\"border-t border-gray-200\"}]}]],[\"$\",\"div\",null,{\"className\":\"flex flex-col gap-4 py-6 px-8 justify-start items-start text-left\",\"children\":[[\"$\",\"div\",null,{\"className\":\"flex flex-col\",\"children\":[[\"$\",\"span\",null,{\"className\":\"text-lg text-gray-500\",\"children\":\"CVPR 2026\"}],[\"$\",\"$L7\",null,{\"href\":\"papers/Interact2Ar\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-5-title-click\",\"children\":[\"$\",\"strong\",null,{\"className\":\"text-xl hover:text-primary transition-all duration-300\",\"children\":\"Interact2Ar: Full-Body Human-Human Interaction Generation via Autoregressive Diffusion Models\\n\"}]}]]}],[\"$\",\"p\",null,{\"children\":\"We introduce Interact2Ar, the first text-conditioned autoregressive diffusion model for generating full-body human-human interactions. Unlike previous approaches that generate entire sequences simultaneously, Interact2Ar utilizes a novel memory strategy and dedicated parallel branches to synthesize high-fidelity motions with detailed hand kinematics. This enables adaptive capabilities, such as temporal composition and real-time adaptation to disturbances. We also introduce robust evaluators and extended metrics to assess full-body interaction quality.\"}],[\"$\",\"div\",null,{\"className\":\"flex flex-wrap gap-4\",\"children\":[[\"$\",\"$L7\",null,{\"href\":\"papers/Interact2Ar\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-5-web-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/web.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"Web\",\"width\":8,\"height\":8}]}],[\"$\",\"$L7\",null,{\"href\":\"https://arxiv.org/abs/2512.19692\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-5-paper-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/paper.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"Paper\",\"width\":8,\"height\":8}]}],[\"$\",\"$L7\",null,{\"href\":\"https://github.com/pabloruizponce/Interact2Ar\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-5-github-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/github.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"GitHub\",\"width\":8,\"height\":8}]}]]}]]}]]}],[\"$\",\"div\",\"4\",{\"className\":\"bg-white rounded-xl shadow border border-gray-200 overflow-hidden undefined\",\"children\":[false,[\"$\",\"div\",null,{\"className\":\"flex flex-col gap-4 py-6 px-8 justify-start items-start text-left\",\"children\":[[\"$\",\"div\",null,{\"className\":\"flex flex-col\",\"children\":[[\"$\",\"span\",null,{\"className\":\"text-lg text-gray-500\",\"children\":\"CVPR 2025\"}],[\"$\",\"$L7\",null,{\"href\":\"papers/MixerMDM\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-4-title-click\",\"children\":[\"$\",\"strong\",null,{\"className\":\"text-xl hover:text-primary transition-all duration-300\",\"children\":\"MixerMDM: Learnable Composition of Human Motion Diffusion Models\"}]}]]}],[\"$\",\"p\",null,{\"children\":\"We introduce MixerMDM, the first learnable approach for combining multiple pre-trained text-conditioned motion diffusion models. Unlike previous static methods, MixerMDM learns a dynamic mixing strategy to adaptively comb
2ine models based on their generated motions and input conditions. This enables fine-grained and consistent individual controllability over human-human interaction motion generation. We also propose a novel evaluation technique that measures mixing quality by assessing individual and interaction alignment.\"}],\"$L15\"]}]]}],\"$L16\",\"$L17\"]}]\n"])</script>
2<script>self.__next_f.push([1,"15:[\"$\",\"div\",null,{\"className\":\"flex flex-wrap gap-4\",\"children\":[[\"$\",\"$L7\",null,{\"href\":\"papers/MixerMDM\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-4-web-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/web.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"Web\",\"width\":8,\"height\":8}]}],[\"$\",\"$L7\",null,{\"href\":\"https://arxiv.org/abs/2504.01019\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-4-paper-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/paper.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"Paper\",\"width\":8,\"height\":8}]}],[\"$\",\"$L7\",null,{\"href\":\"https://github.com/pabloruizponce/MixerMDM\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-4-github-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/github.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"GitHub\",\"width\":8,\"height\":8}]}]]}]\n"])</script>
2<script>self.__next_f.push([1,"16:[\"$\",\"div\",\"1\",{\"className\":\"bg-white rounded-xl shadow border border-gray-200 overflow-hidden undefined\",\"children\":[false,[\"$\",\"div\",null,{\"className\":\"flex flex-col gap-4 py-6 px-8 justify-start items-start text-left\",\"children\":[[\"$\",\"div\",null,{\"className\":\"flex flex-col\",\"children\":[[\"$\",\"span\",null,{\"className\":\"text-lg text-gray-500\",\"children\":\"CVPR 2024 HuMoGen Workshop\"}],[\"$\",\"$L7\",null,{\"href\":\"https://pabloruizponce.github.io/in2IN/\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-1-title-click\",\"children\":[\"$\",\"strong\",null,{\"className\":\"text-xl hover:text-primary transition-all duration-300\",\"children\":\"in2IN: Leveraging individual Information to Generate Human INteractions\"}]}]]}],[\"$\",\"p\",null,{\"children\":\"We present in2IN, a diffusion model for human-human motion generation that uses both overall and individual action descriptions to improve the realism and diversity of interactions. By augmenting the InterHuman dataset with individual textual annotations using a large language model, in2IN achieves state-of-the-art results. To further boost intra-person diversity, we propose DualMDM, a model composition method that combines motions from in2IN with a single-person model trained on HumanML3D. Together, they enhance control over individual behavior while preserving coherent interaction dynamics.\"}],[\"$\",\"div\",null,{\"className\":\"flex flex-wrap gap-4\",\"children\":[[\"$\",\"$L7\",null,{\"href\":\"https://pabloruizponce.github.io/in2IN/\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-1-web-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/web.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"Web\",\"width\":8,\"height\":8}]}],[\"$\",\"$L7\",null,{\"href\":\"https://arxiv.org/abs/2404.09988\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-1-paper-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/paper.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"Paper\",\"width\":8,\"height\":8}]}],[\"$\",\"$L7\",null,{\"href\":\"https://github.com/pabloruizponce/in2IN\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-1-github-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/github.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"GitHub\",\"width\":8,\"height\":8}]}]]}]]}]]}]\n"])</script>
2<script>self.__next_f.push([1,"17:[\"$\",\"div\",\"2\",{\"className\":\"bg-white rounded-xl shadow border border-gray-200 overflow-hidden undefined\",\"children\":[false,[\"$\",\"div\",null,{\"className\":\"flex flex-col gap-4 py-6 px-8 justify-start items-start text-left\",\"children\":[[\"$\",\"div\",null,{\"className\":\"flex flex-col\",\"children\":[[\"$\",\"span\",null,{\"className\":\"text-lg text-gray-500\",\"children\":\"UPC Commons\"}],[\"$\",\"$L7\",null,{\"href\":\"https://upcommons.upc.edu/handle/2117/408075\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-2-title-click\",\"children\":[\"$\",\"strong\",null,{\"className\":\"text-xl hover:text-primary transition-all duration-300\",\"children\":\"Text-driven multi-human motion generation\"}]}]]}],[\"$\",\"p\",null,{\"children\":\"In this thesis, we introduce a novel Diffusion Model incorporating a Transformer-based architecture. This model is conditioned using textual descriptions of both the motion interactions and the individual motions within these interactions. By focusing on the individual components of the interaction, our method achieves more precise conditioning in the generation of these specific motions. Concurrently, the textual descriptions of the overall interaction enable our model to effectively capture the interplay between individual motions.\"}],[\"$\",\"div\",null,{\"className\":\"flex flex-wrap gap-4\",\"children\":[[\"$\",\"$L7\",null,{\"href\":\"https://upcommons.upc.edu/handle/2117/408075\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-2-web-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/web.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"Web\",\"width\":8,\"height\":8}]}],[\"$\",\"$L7\",null,{\"href\":\"https://upcommons.upc.edu/handle/2117/408075\",\"target\":\"_blank\",\"rel\":\"noopener noreferrer\",\"data-umami-event\":\"paper-card-2-paper-click\",\"children\":[\"$\",\"$L14\",null,{\"src\":\"icons/paper.svg\",\"className\":\"w-8 h-8 hover:scale-110 transition-all duration-300\",\"alt\":\"Paper\",\"width\":8,\"height\":8}]}],null]}]]}]]}]\n"])</script>
2</body></html>

Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.