https://lxk-221.github.io/tags/tactile-sensing/Tactile SensingHugo Blox Builder (https://hugoblox.com)en-usThu, 05 Feb 2026 00:00:00 +0000https://lxk-221.github.io/media/icon_hu68170e94a17a2a43d6dcb45cf0e8e589_3079_512x512_fill_lanczos_center_3.pnghttps://lxk-221.github.io/tags/tactile-sensing/https://lxk-221.github.io/publication/deco/Thu, 05 Feb 2026 00:00:00 +0000https://lxk-221.github.io/publication/deco/<div class="flex px-4 py-3 mb-6 rounded-md bg-primary-100 dark:bg-primary-900"> <span class="pr-3 pt-1 text-primary-600 dark:text-primary-300"> <svg height="24" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path fill="none" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="1.5" d="m11.25 11.25l.041-.02a.75.75 0 0 1 1.063.852l-.708 2.836a.75.75 0 0 0 1.063.853l.041-.021M21 12a9 9 0 1 1-18 0a9 9 0 0 1 18 0m-9-3.75h.008v.008H12z"/></svg> </span> <span class="dark:text-neutral-300">Released on arXiv:</span> </div> <p>DECO is a decoupled multimodal diffusion transformer for bimanual dexterous manipulation that:</p> <ul> <li><strong>Disentangles multimodal inputs</strong> (vision, proprioception, tactile) through specialized conditioning pathways</li> <li><strong>Features a lightweight tactile adapter</strong> for parameter-efficient injection of tactile signals</li> <li><strong>Achieves 72.25% success rate</strong> with 21% improvement over baseline</li> <li><strong>Introduces DECO-50 dataset</strong> with 50 hours of data and over 5M frames</li> </ul>