Metraipolis › archive › dc_53fd3a417d3d
AcrossDEX 1.0: From Human Motion Pretraining to Bimanual Dexterous Manipulation across Hand Embodiments — arXiv:2610.40210
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.40210
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-01T02:45:48.243Z
- SHA-256 of the archived text
d7954a0823b23105ac5dd8f7f67f6c94024aca6246242336faf769e95b478795- Stable URL of this page
https://metraipolis.com/documento/dc_53fd3a417d3d
How to cite
AcrossDEX 1.0: From Human Motion Pretraining to Bimanual Dexterous Manipulation across Hand Embodiments — arXiv:2610.40210 — arXiv (en), https://arxiv.org/abs/2610.40210 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_53fd3a417d3d (Metraipolis archive dc_53fd3a417d3d, sha256 d7954a0823b23105ac5dd8f7f67f6c94024aca6246242336faf769e95b478795)
BibTeX
@misc{dc53fd3a417d3d,
title = {AcrossDEX 1.0: From Human Motion Pretraining to Bimanual Dexterous Manipulation across Hand Embodiments — arXiv:2610.40210},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.40210},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_53fd3a417d3d; sha256 d7954a0823b23105ac5dd8f7f67f6c94024aca6246242336faf769e95b478795},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40210. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Bimanual dexterous manipulation requires continuous coordination of wrist motion, finger contact, and object motion. Human videos provide rich manipulation experience. Using this experience for robot control requires aligning human poses with robot commands and accounting for embodiment-specific kinematics and contact dynamics. We present AcrossDEX 1.0, a learning system that connects human motion pretraining, simulation-based expert learning, robot demonstration collection, and visual closed-loop execution. Pretraining on HOT3D Aria predicts one second of bimanual motion from five images and thirty annotated hand states over the preceding second. Robot adaptation preserves the visual encoder, temporal backbone, and wrist parameters. Shared wrist representations, separate finger fields, and validity masks support both six-channel Inspire hands and sixteen-joint Allegro hands. The robot policy predicts sixty control targets from frontal images, command history, and task conditions. It executes a prefix and replans from updated observations. We compare direct regression, diffusion, and flow matching on lid opening and return and closed-lid object reorientation. Direct regression achieves success rates of 40.6\% for opening and return, 100.0\% for Inspire reorientation, and 93.3\% for Allegro reorientation. Flow matching achieves 90.0\% on Allegro reorientation. Paired comparisons further show that constraining replanning with the previous plan improves the continuous execution of generative policies. These results demonstrate a common adaptation pipeline for two hand embodiments and highlight the roles of contact coordination and action-chunk continuity. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2610.40210 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-01T02:16:58.679Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).