1<!DOCTYPE html> <html lang="en"> <head> <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> <meta charset="utf-8"> <meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"> <meta http-equiv="X-UA-Compatible" content="IE=edge"> <title> publications | Yutang Lin </title> <meta name="author" content="Yutang Lin"> <meta name="description" content="A comprehensive collection of academic publications organized by type and sorted chronologically."> <meta name="keywords" content="jekyll, jekyll-theme, academic-website, portfolio-website"> <link rel="stylesheet" href="/assets/css/bootstrap.min.css?a4b3f509e79c54a512b890d73235ef04"> <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/css/mdb.min.css" integrity="sha256-jpjYvU3G3N6nrrBwXJoVEYI/0zw8htfFnhT9ljN3JJw=" crossorigin="anonymous"> <link defer rel="stylesheet" href="/assets/css/academicons.min.css?f0b7046b84e425c55f3463ac249818f5"> <link defer rel="stylesheet" href="/assets/css/scholar-icons.css?62b2ac103a88034e6882a5be5f3e2772"> <link defer rel="stylesheet" type="text/css" href="https://fonts.googleapis.com/css2?family=Fraunces:ital,opsz,wght@0,9..144,300..900;1,9..144,300..700&
1family=Inter:wght@300;400;500;600;700;800&family=JetBrains+Mono:wght@400;500&family=Roboto:wght@300;400;500;700&family=Roboto+Slab:wght@100;300;400;500;700&family=Material+Icons&display=swap"> <link defer rel="stylesheet" href="/assets/css/jekyll-pygments-themes-github.css?591dab5a4e56573bf4ef7fd332894c99" media="" id="highlight_theme_light"> <link rel="shortcut icon" href="data:image/svg+xml,<svg%20xmlns=%22http://www.w3.org/2000/svg%22%20viewBox=%220%200%20100%20100%22><text%20y=%22.9em%22%20font-size=%2290%22>%F0%9F%A4%AA</text></svg>"> <link rel="stylesheet" href="/assets/css/main.css?d41d8cd98f00b204e9800998ecf8427e"> <link rel="canonical" href="https://yutang-lin.github.io/publications/">
1<script src="/assets/js/theme.js?4cd6edf741c75aac92971c2b7b959939"></script>
1 <link defer rel="stylesheet" href="/assets/css/jekyll-pygments-themes-native.css?5847e5ed4a4568527aa6cfab446049ca" media="none" id="highlight_theme_dark">
1<script> 2 initTheme(); 3 </script>
3 </head> <body class="fixed-top-nav "> <header> <nav id="navbar" class="navbar navbar-light navbar-expand-sm fixed-top" role="navigation"> <div class="container"> <a class="navbar-brand title font-weight-lighter" href="/"> <span class="font-weight-bold">Yutang</span> Lin </a> <button class="navbar-toggler collapsed ml-auto" type="button" data-toggle="collapse" data-target="#navbarNav" aria-controls="navbarNav" aria-expanded="false" aria-label="Toggle navigation"> <span class="sr-only">Toggle navigation</span> <span class="icon-bar top-bar"></span> <span class="icon-bar middle-bar"></span> <span class="icon-bar bottom-bar"></span> </button> <div class="collapse navbar-collapse text-right" id="navbarNav"> <ul class="navbar-nav ml-auto flex-nowrap"> <li class="nav-item "> <a class="nav-link" href="/">about </a> </li> <li class="nav-item "> <a class="nav-link" href="/blog/">blog </a> </li> <li class="nav-item active"> <a class="nav-link" href="/publications/">publications <span class="sr-only">(current)</span> </a> </li> <li class="nav-item "> <a class="nav-link" href="/repositories/">repositories </a> </li> <li class="nav-item "> <a class="nav-link" href="/cv/">cv </a> </li> <li class="nav-item dropdown "> <a class="nav-link dropdown-toggle" href="#" id="navbarDropdown" role="button" data-toggle="dropdown" aria-haspopup="true" aria-expanded="false">submenus </a> <div class="dropdown-menu dropdown-menu-right" aria-labelledby="navbarDropdown"> <a class="dropdown-item " href="/books/">bookshelf</a> </div> </li> <li class="nav-item"> <button id="search-toggle" title="Search" onclick="openSearchModal()"> <span class="nav-link">ctrl k <i class="ti ti-search"></i></span> </button> </li> <li class="toggle-container"> <button id="light-toggle" title="Change theme"> <i class="ti ti-sun-moon" id="light-toggle-system"></i> <i class="ti ti-moon-filled" id="light-toggle-dark"></i> <i class="ti ti-sun-filled" id="light-toggle-light"></i> </button> </li> </ul> </div> </div> </nav> <progress id="progress" value="0"> <div class="progress-container"> <span class="progress-bar"></span> </div> </progress> </header> <div class="container mt-5" role="main"> <div class="post"> <header class="post-header"> <h1 class="post-title">publications</h1> <p class="post-description">A comprehensive collection of academic publications organized by type and sorted chronologically.</p> </header> <article>
3<script src="/assets/js/bibsearch.js?1bc438ca9037884cc579601c09afd847" type="module"></script>
3 <p><input type="text" id="bibsearch" spellcheck="false" autocomplete="off" class="search bibsearch-form-input" placeholder="Type to filter"></p> <div class="publications"> <h2 class="bibliography">Pre-Prints</h2> <ol class="bibliography"> <li> <div class="row"> <div class="col col-sm-2 abbr"> <abbr class="badge rounded w-100" style="background-color:#9aa0a6"> <div>Coming Soon</div> </abbr> <figure> <picture> <img src="/assets/img/publication_preview/question_mark.png" class="preview z-depth-1 rounded" width="100%" height="auto" alt="question_mark.png" data-zoomable="" loading="eager" onerror="this.onerror=null; $('.responsive-img-srcset').remove();"> </picture> </figure> </div> <div id="coming1" class="col-sm-8"> <div class="title">Coming Soon</div> <div class="author"> <em>Yutang Lin</em>, ???, and ??? </div> <div class="periodical"> <em>Coming Soon</em>, Sep 2026 </div> <div class="periodical"> </div> <div class="links"> </div> </div> </div> </li> <li> <div class="row"> <div class="col col-sm-2 abbr"> <abbr class="badge rounded w-100" style="background-color:#b31b1b"> <a href="https://arxiv.org/" rel="external nofollow noopener" target="_blank">arXiv</a> </abbr> <figure> <picture> <img src="/assets/img/publication_preview/omniclone.gif" class="preview z-depth-1 rounded" width="100%" height="auto" alt="omniclone.gif" data-zoomable="" loading="eager" onerror="this.onerror=null; $('.responsive-img-srcset').remove();"> </picture> </figure> </div> <div id="omniclone" class="col-sm-8"> <div class="title">OmniClone: Engineering a Robust, All-Rounder Whole-Body Humanoid Teleoperation System</div> <div class="author"> Yixuan Li<sup>*</sup>, Le Ma<sup>*</sup>, <em>Yutang Lin<sup>*</sup></em>, and <span class="more-authors" title="click to view 8 more authors" onclick=" var element=$(this); element.attr('title', ''); var more_authors_text=element.text() == '8 more authors' ? 'Yushi Du, Mengya Liu, Kaizhe Hu, Jieming Cui, Yixin Zhu<sup>â </sup>, Wei Liang<sup>â </sup>, Baoxiong Jia<sup>â </sup>, Siyuan Huang<sup>â </sup>' : '8 more authors'; var cursorPosition=0; var textAdder=setInterval(function(){ element.html(more_authors_text.substring(0, cursorPosition + 1)); if (++cursorPosition == more_authors_text.length){ clearInterval(textAdder); } }, '10'); ">8 more authors</span> </div> <div class="periodical"> <em>arXiv</em>, Mar 2026 </div> <div class="periodical"> </div> <div class="links"> <a class="abstract btn btn-sm z-depth-0" role="button">Abs</a> <a href="http://arxiv.org/abs/2603.14327" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">arXiv</a> <a href="/assets/pdf/omniclone.pdf" class="btn btn-sm z-depth-0" role="button">PDF</a> <a href="https://vimeo.com/1171516179" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Video</a> <a href="https://github.com/yixxuan-li/OmniClone" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Code</a> <a href="https://omniclone.github.io/" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Website</a> </div> <div class="abstract hidden"> <p>OmniClone, a whole-body humanoid teleoperation system that achieves robust, high-fidelity multi-skill control on a single consumer GPU with modest data. We introduce OmniBench, a diagnostic benchmark that disentangles motion regimes and reveals failure modes overlooked by prior aggregate metrics, guiding an optimized training data recipe and system design. Through subject-agnostic retargeting and robust communication, OmniClone reduces MPJPE by over 66% while requiring significantly fewer resources. The system supports unified, control-source-agnostic operationâenabling teleoperation, motion playback, and VLA integrationâand generalizes across operators with diverse body proportions, offering a scalable and practical foundation for humanoid control and learning.</p> </div> </div> </div> </li> </ol> <h2 class="bibliography">Journal Articles</h2> <ol class="bibliography"><li> <div class="row"> <div class="col col-sm-2 abbr"> <abbr class="badge rounded w-100" style="background-color:#16a34a"> <a href="https://link.springer.com/journal/11263" rel="external nofollow noopener" target="_blank">IJCV</a> </abbr> <figure> <picture> <img src="/assets/img/publication_preview/alphachimp.gif" class="preview z-depth-1 rounded" width="100%" height="auto" alt="alphachimp.gif" data-zoomable="" loading="eager" onerror="this.onerror=null; $('.responsive-img-srcset').remove();"> </picture> </figure> </div> <div id="alphachimp" class="col-sm-8"> <div class="title">AlphaChimp: Tracking and Behavior Recognition of Chimpanzees</div> <div class="author"> Xiaoxuan Ma<sup>*</sup>, <em>Yutang Lin<sup>*</sup></em>, Yuan Xu, and <span class="more-authors" title="click to view 6 more authors" onclick=" var element=$(this); element.attr('title', ''); var more_authors_text=element.text() == '6 more authors' ? 'Stephan P. Kaufhold, Jack Terwilliger, Andres Meza, Yixin Zhu<sup>â </sup>, Federico Rossano<sup>â </sup>, Yizhou Wang<sup>â </sup>' : '6 more authors'; var cursorPosition=0; var textAdder=setInterval(function(){ element.html(more_authors_text.substring(0, cursorPosition + 1)); if (++cursorPosition == more_authors_text.length){ clearInterval(textAdder); } }, '10'); ">6 more authors</span> </div> <div class="periodical"> <em>IJCV</em>, Oct 2024 </div> <div class="periodical"> </div> <div class="links"> <a class="abstract btn btn-sm z-depth-0" role="button">Abs</a> <a href="http://arxiv.org/abs/2410.17136" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">arXiv</a> <a href="/assets/pdf/alphachimp.pdf" class="btn btn-sm z-depth-0" role="button">PDF</a> <a href="https://drive.google.com/file/d/15G950d5jPrdssjEQd6GB2jEU1Xcd8EVi/view" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Video</a> <a href="https://github.com/ShirleyMaxx/AlphaChimp" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Code</a> <a href="https://sites.google.com/view/alphachimp/home/" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Website</a> </div> <div class="abstract hidden"> <p>Understanding non-human primate behavior is crucial for improving animal welfare, modeling social behavior, and gaining insights into both distinctly human and shared behaviors. Despite recent advances in computer vision, automated analysis of primate behavior remains challenging due to the complexity of their social interactions and the lack of specialized algorithms. Existing methods often struggle with the nuanced behaviors and frequent occlusions characteristic of primate social dynamics. This study aims to develop an effective method for automated detection, tracking, and recognition of chimpanzee behaviors in video footage. Here we show that our proposed method, AlphaChimp, an end-to-e
3nd approach that simultaneously detects chimpanzee positions and estimates behavior categories from videos, significantly outperforms existing methods in behavior recognition. AlphaChimp achieves approximately 10% higher tracking accuracy and a 20% improvement in behavior recognition compared to state-of-the-art methods, particularly excelling in the recognition of social behaviors. This superior performance stems from AlphaChimpâs innovative architecture, which integrates temporal feature fusion with a Transformer-based self-attention mechanism, enabling more effective capture and interpretation of complex social interactions among chimpanzees. Our approach bridges the gap between computer vision and primatology, enhancing technical capabilities and deepening our understanding of primate communication and sociality. We release our code and models and hope this will facilitate future research in animal social dynamics. This work contributes to ethology, cognitive science, and artificial intelligence, offering new perspectives on social intelligence.</p> </div> </div> </div> </li></ol> <h2 class="bibliography">Conference Articles</h2> <ol class="bibliography"> <li> <div class="row"> <div class="col col-sm-2 abbr"> <abbr class="badge rounded w-100">NeurIPS</abbr> <figure> <picture> <img src="/assets/img/publication_preview/lessmimic.gif" class="preview z-depth-1 rounded" width="100%" height="auto" alt="lessmimic.gif" data-zoomable="" loading="eager" onerror="this.onerror=null; $('.responsive-img-srcset').remove();"> </picture> </figure> </div> <div id="lessmimic" class="col-sm-8"> <div class="title">LessMimic: Long-Horizon Humanoid Interaction with Unified Distance Field Representations</div> <div class="author"> <em>Yutang Lin<sup>*</sup></em>, Jieming Cui<sup>*</sup>, Yixuan Li, and <span class="more-authors" title="click to view 3 more authors" onclick=" var element=$(this); element.attr('title', ''); var more_authors_text=element.text() == '3 more authors' ? 'Baoxiong Jia<sup>â </sup>, Yixin Zhu<sup>â </sup>, Siyuan Huang<sup>â </sup>' : '3 more authors'; var cursorPosition=0; var textAdder=setInterval(function(){ element.html(more_authors_text.substring(0, cursorPosition + 1)); if (++cursorPosition == more_authors_text.length){ clearInterval(textAdder); } }, '10'); ">3 more authors</span> </div> <div class="periodical"> <em>NeurIPS</em>, Feb 2026 </div> <div class="periodical"> </div> <div class="links"> <a class="abstract btn btn-sm z-depth-0" role="button">Abs</a> <a href="http://arxiv.org/abs/2602.21723" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">arXiv</a> <a href="/assets/pdf/lessmimic.pdf" class="btn btn-sm z-depth-0" role="button">PDF</a> <a href="https://vimeo.com/1168055451" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Video</a> <a href="https://github.com/Yutang-Lin/LessMimic" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Code</a> <a href="https://lessmimic.github.io/" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Website</a> </div> <div class="abstract hidden"> <p>Humanoid robots that autonomously interact with physical environments over extended horizons represent a central goal of embodied intelligence. Existing approaches rely on reference motions or task-specific rewards, tightly coupling policies to particular object geometries and precluding multi-skill generalization within a single framework. A unified interaction representation enabling reference-free inference, geometric generalization, and long-horizon skill composition within one policy remains an open challenge. Here we show that Distance Field (DF) provides such a representation: LessMimic conditions a single whole-body policy on DF-derived geometric cuesâsurface distances, gradients, and velocity decompositionsâremoving the need for motion references, with interaction latents encoded via a Variational Autoencoder (VAE) and post-trained using Adversarial Interaction Prior (AIP)-derived Reinforcement Learning (RL). Through DAgger-style distillation that aligns DF latents with egocentric depth features, LessMimic further transfers seamlessly to vision-only deployment without motion capture infrastru
3cture.</p> </div> </div> </div> </li> <li> <div class="row"> <div class="col col-sm-2 abbr"> <abbr class="badge rounded w-100" style="background-color:#1e3a8a"> <a href="https://openreview.net/forum?id=AC1XQjy8Sa" rel="external nofollow noopener" target="_blank">NeurIPS-W</a> </abbr> <figure> <picture> <img src="/assets/img/publication_preview/gapo.png" class="preview z-depth-1 rounded" width="100%" height="auto" alt="gapo.png" data-zoomable="" loading="eager" onerror="this.onerror=null; $('.responsive-img-srcset').remove();"> </picture> </figure> </div> <div id="gaponet" class="col-sm-8"> <div class="title">GapONet: Nonlinear Operator Learning for Bridging the Humanoid Sim-to-Real Gap</div> <div class="author"> Jieming Cui<sup>*</sup>, Zhenghao Qi<sup>*</sup>, <em>Yutang Lin<sup>*</sup></em>, and <span class="more-authors" title="click to view 6 more authors" onclick=" var element=$(this); element.attr('title', ''); var more_authors_text=element.text() == '6 more authors' ? 'Tengyu Liu, Yuntian Hu, Bin He, Ruihua Zhang<sup>â </sup>, Siyuan Huang<sup>â </sup>, Yixin Zhu<sup>â </sup>' : '6 more authors'; var cursorPosition=0; var textAdder=setInterval(function(){ element.html(more_authors_text.substring(0, cursorPosition + 1)); if (++cursorPosition == more_authors_text.length){ clearInterval(textAdder); } }, '10'); ">6 more authors</span> </div> <div class="periodical"> <em>NeurIPS-W</em>, Dec 2025 </div> <div class="periodical"> </div> <div class="links"> <a class="abstract btn btn-sm z-depth-0" role="button">Abs</a> <a href="https://openreview.net/forum?id=AC1XQjy8Sa" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">HTML</a> </div> <div class="abstract hidden"> <p>Sim-to-real transfer remains a central challenge for humanoid robots, where dynamics mismatches between simulation and hardware degrade control performance. We propose GapONet, a payload-conditioned nonlinear operator that maps simulation context functions to residual actions for hardware. We compile TWINS, a dataset of over 11,298 motion sequences, perform payload-aware system identification, and train GapONet to predict compensatory actions in real time. Our method keeps large gaps below 0.09% when tracking unseen motions, substantially outperforming competing approaches.</p> </div> </div> </div> </li> <li> <div class="row"> <div class="col col-sm-2 abbr"> <abbr class="badge rounded w-100" style="background-color:#2563eb"> <a href="https://icra.ieee.org/" rel="external nofollow noopener" target="_blank">ICRA</a> </abbr> <figure> <picture> <img src="/assets/img/publication_preview/cola.gif" class="preview z-depth-1 rounded" width="100%" height="auto" alt="cola.gif" data-zoomable="" loading="eager" onerror="this.onerror=null; $('.responsive-img-srcset').remove();"> </picture> </figure> </div> <div id="cola" class="col-sm-8"> <div class="title">COLA: Learning Human-Humanoid Coordination for Collaborative Object Carrying</div> <div class="author"> Yushi Du<sup>*</sup>, Yixuan Li<sup>*</sup>, Baoxiong Jia<sup>*</sup>, <em>Yutang Lin</em>, and <span class="more-authors" title="click to view 4 more authors" onclick=" var element=$(this); element.attr('title', ''); var more_authors_text=element.text() == '4 more authors' ? 'Pei Zhou, Wei Liang<sup>â </sup>, Yanchao Yang<sup>â </sup>, Siyuan Huang<sup>â </sup>' : '4 more authors'; var cursorPosition=0; var textAdder=setInterval(function(){ element.html(more_authors_text.substring(0, cursorPosition + 1)); if (++cursorPosition == more_authors_text.length){ clearInterval(textAdder); } }, '10'); ">4 more authors</span> </div> <div class="periodical"> <em>ICRA</em>, Oct 2025 </div> <div class="periodical"> </div> <div class="links"> <a class="abstract btn btn-sm z-depth-0" role="button">Abs</a> <a href="http://arxiv.org/abs/2510.14293" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">arXiv</a> <a href="/assets/pdf/cola.pdf" class="btn btn-sm z-depth-0" role="button">PDF</a> <a href="https://yushi-du.github.io/COLA/static/videos/No_method.mp4" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Video</a> <a href="https://github.com/Yushi-Du/COLA_Code" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Code</a> <a href="https://yushi-du.github.io/COLA/" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Website</a> </div> <div class="abstract hidden"> <p>Human-humanoid collaboration shows significant promise for applications in healthcare, domestic assistance, and manufacturing. While compliant robot-human collaboration has been extensively developed for robotic arms, enabling compliant human-humanoid collaboration remains largely unexplored due to humanoidsâ complex whole-body dynamics. In this paper, we propose a proprioception-only reinforcement learning approach, COLA, that c
3ombines leader and follower behaviors within a single policy. The model is trained in a closed-loop environment with dynamic object interactions to predict object motion patterns and human intentions implicitly, enabling compliant collaboration to maintain load balance through coordinated trajectory planning.</p> </div> </div> </div> </li> <li> <div class="row"> <div class="col col-sm-2 abbr"> <abbr class="badge rounded w-100" style="background-color:#f59e0b"> <a href="https://www.corl.org/" rel="external nofollow noopener" target="_blank">CoRL</a> </abbr> <figure> <picture> <img src="/assets/img/publication_preview/clone.gif" class="preview z-depth-1 rounded" width="100%" height="auto" alt="clone.gif" data-zoomable="" loading="eager" onerror="this.onerror=null; $('.responsive-img-srcset').remove();"> </picture> </figure> </div> <div id="clone" class="col-sm-8"> <div class="title">CLONE: Holistic Closed-Loop Whole-Body Teleoperation for Long-Horizon Humanoid Control</div> <div class="author"> Yixuan Li<sup>*</sup>, <em>Yutang Lin<sup>*</sup></em>, Jieming Cui, and <span class="more-authors" title="click to view 4 more authors" onclick=" var element=$(this); element.attr('title', ''); var more_authors_text=element.text() == '4 more authors' ? 'Tengyu Liu, Wei Liang<sup>â </sup>, Yixin Zhu<sup>â </sup>, Siyuan Huang<sup>â </sup>' : '4 more authors'; var cursorPosition=0; var textAdder=setInterval(function(){ element.html(more_authors_text.substring(0, cursorPosition + 1)); if (++cursorPosition == more_authors_text.length){ clearInterval(textAdder); } }, '10'); ">4 more authors</span> </div> <div class="periodical"> <em>CoRL</em>, Jun 2025 </div> <div class="periodical"> </div> <div class="links"> <a class="abstract btn btn-sm z-depth-0" role="button">Abs</a> <a href="http://arxiv.org/abs/2506.08931" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">arXiv</a> <a href="/assets/pdf/clone.pdf" class="btn btn-sm z-depth-0" role="button">PDF</a> <a href="https://humanoid-clone.github.io/resources/overview_v0.1.mp4" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Video</a> <a href="https://github.com/humanoid-clone/CLONE" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Code</a> <a href="https://humanoid-clone.github.io/" class="btn btn-sm z-depth-0" role="button" rel="external nofollow noopener" target="_blank">Website</a> </div> <div class="abstract hidden"> <p>Humanoid teleoperation plays a vital role in demonstrating and collecting data for complex humanoid-scene interactions. However, current teleoperation systems face critical limitations: they decouple upper- and lower-body control to maintain stability, restricting natural coordination, and operate open-loop without real-time position feedback, leading to accumulated drift. The fundamental challenge is achieving precise, coordinated whole-body teleoperation over extended durations while maintaining accurate global positioning. Here we show that an MoE-based teleoperation system, CLONE, with closed-loop error correction enables unprecedented whole-body teleoperation fidelity, maintaining minimal positional drift over long-range trajectories using only head and hand tracking from an MR headset. Unlike previous methods that either sacrifice coordination for stability or suffer from unbounded drift, CLONE learns diverse motion skills while preventing tracking error accumulation through real-time feedback, enabling complex coordinated movements such as "picking up objects from the ground." These results establish a new milestone for whole-body humanoid teleoperation for long-horizon humanoid-scene interaction tasks.</p> </div> </div> </div> </li> </ol> </div> </article> </div> </div> <footer class="fixed-bottom" role="contentinfo"> <div class="container mt-0"> © Copyright 2026 Yutang Lin. Powered by <a href="https://jekyllrb.com/" target="_blank" rel="external nofollow noopener">Jekyll</a> with <a href="https://github.com/alshedivat/al-folio" rel="external nofollow noopener" target="_blank">al-folio</a> theme. Hosted by <a href="https://pages.github.com/" target="_blank" rel="external nofollow noopener">GitHub Pages</a>. Photos from <a href="https://unsplash.com" target="_blank" rel="external nofollow noopener">Unsplash</a>. </div> </footer>
3<script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/jquery.min.js" integrity="sha256-/xUj+3OJU5yExlq6GSYGSHk7tPXikynS7ogEvDej/m4=" crossorigin="anonymous"></script>
3
3<script src="/assets/js/bootstrap.bundle.min.js"></script>
3
3<script src="https://cdn.jsdelivr.net/npm/[email protected]/js/mdb.min.js" integrity="sha256-NdbiivsvWt7VYCt6hYNT3h/th9vSTL4EDWeGs5SN3DA=" crossorigin="anonymous"></script>
3
3<script defer src="https://cdn.jsdelivr.net/npm/[email protected]/dist/masonry.pkgd.min.js" integrity="sha256-Nn1q/fx0H7SNLZMQ5Hw5JLaTRZp0yILA/FRexe19VdI=" crossorigin="anonymous"></script>
3
3<script defer src="https://cdn.jsdelivr.net/npm/[email protected]/imagesloaded.pkgd.min.js" integrity="sha256-htrLFfZJ6v5udOG+3kNLINIKh2gvoKqwEhHYfTTMICc=" crossorigin="anonymous"></script>
3
3<script defer src="/assets/js/masonry.js?a0db7e5d5c70cc3252b3138b0c91dcaf" type="text/javascript"></script>
3
3<script defer src="https://cdn.jsdelivr.net/npm/[email protected]/dist/medium-zoom.min.js" integrity="sha256-ZgMyDAIYDYGxbcpJcfUnYwNevG/xi9OHKaR/8GK+jWc=" crossorigin="anonymous"></script>
3
3<script defer src="/assets/js/zoom.js?85ddb88934d28b74e78031fd54cf8308"></script>
3
3<script src="/assets/js/no_defer.js?2781658a0a2b13ed609542042a859126"></script>
3
3<script defer src="/assets/js/common.js?5d7a7666d5209ade57efdb75b71b032c"></script>
3
3<script defer src="/assets/js/copy_code.js?c8a01c11a92744d44b093fc3bda915df" type="text/javascript"></script>
3
3<script defer src="/assets/js/jupyter_new_tab.js?d9f17b6adc2311cbabd747f4538bb15f"></script>
3
3<script async src="https://d1bxh8uas1mnw7.cloudfront.net/assets/embed.js"></script>
3
3<script async src="https://badge.dimensions.ai/badge.js"></script>
3
3<script defer type="text/javascript" id="MathJax-script" src="https://cdn.jsdelivr.net/npm/[email protected]/es5/tex-mml-chtml.js" integrity="sha256-MASABpB4tYktI2Oitl4t+78w/lyA+D7b/s9GEP0JOGI=" crossorigin="anonymous"></script>
3
3<script src="/assets/js/mathjax-setup.js?a5bb4e6a542c546dd929b24b8b236dfd"></script>
3
3<script defer src="https://cdnjs.cloudflare.com/polyfill/v3/polyfill.min.js?features=es6" crossorigin="anonymous"></script>
3
3<script defer src="/assets/js/progress-bar.js?2f30e0e6801ea8f5036fa66e1ab0a71a" type="text/javascript"></script>
3
3<script src="/assets/js/vanilla-back-to-top.min.js?f40d453793ff4f64e238e420181a1d17"></script>
3
3<script> 4 addBackToTop(); 5 </script>
5
5<script type="module" src="/assets/js/search/ninja-keys.min.js?a3446f084dcaecc5f75aa1757d087dcf"></script>
5 <ninja-keys hidebreadcrumbs noautoloadmdicons placeholder="Type to start searching"></ninja-keys>
5<script src="/assets/js/search-setup.js?6c304f7b1992d4b60f7a07956e52f04a"></script>
5
5<script src="/assets/js/search-data.js"></script>
5
5<script src="/assets/js/shortcut-key.js?6f508d74becd347268a7f822bca7309d"></script>
5 </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.