Metraipolis › archive › dc_082e4e35bd0e

MachEmbodied-U0: Unified Understanding and Generation Model for Embodied Intelligence — arXiv:2609.25627

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.25627
License (as archived): CC0-1.0
Archived: 2026-10-08T16:39:54.419Z
SHA-256 of the archived text: ac1e71c76d3d26ca9b76b4a83a916194c178bc8fdf3c85836df2060c89064594
Stable URL of this page: https://metraipolis.com/documento/dc_082e4e35bd0e

How to cite

MachEmbodied-U0: Unified Understanding and Generation Model for Embodied Intelligence — arXiv:2609.25627 — arXiv (en), https://arxiv.org/abs/2609.25627 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_082e4e35bd0e (Metraipolis archive dc_082e4e35bd0e, sha256 ac1e71c76d3d26ca9b76b4a83a916194c178bc8fdf3c85836df2060c89064594)

BibTeX

@misc{dc082e4e35bd0e,
  title = {MachEmbodied-U0: Unified Understanding and Generation Model for Embodied Intelligence — arXiv:2609.25627},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.25627},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_082e4e35bd0e; sha256 ac1e71c76d3d26ca9b76b4a83a916194c178bc8fdf3c85836df2060c89064594},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.25627. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

General-purpose robot control requires models to understand task intent, identify where to interact, capture how the scene evolves, and generate precise actions. Vision-language-action models provide strong semantic priors but typically do not explicitly model scene dynamics, while world-action models couple visual prediction with control without necessarily exposing the task-relevant semantic and spatial structure needed for fine-grained manipulation. We present MachEmbodied-U0 (ME-U0), a unified embodied foundation model connecting understanding and generation experts through a Mixture-of-Transformers architecture. Subtask prediction and affordance grounding guide joint visual-dynamics and action generation via flow matching. Visual dynamics encompass future RGB, depth, surface normals, and optical flow, providing complementary supervision for appearance, geometry, and motion. Multi-rate Rotary Position Encoding (MRPE) aligns visual dynamics with fine-grained control. We pretrain ME-U0 on approximately 4,200 hours of curated demonstrations from robotic datasets and egocentric datasets. Using only the supervision natively available in each downstream benchmark, ME-U0 achieves an average score of 17.66 on the RoboDojo simulation benchmark and average success rates of 99.0\% and 82.5\% on LIBERO and LIBERO-Plus, respectively. We additionally validate ME-U0 on real-world robotic manipulation tasks, demonstrating its effectiveness beyond simulation. Without corresponding downstream supervision, ME-U0 further demonstrates zero-shot subtask prediction, affordance grounding, and visual dynamics on simulated and real-world observations. Overall, ME-U0 combines competitive downstream control performance with transferable task-grounding and visual-dynamics capabilities across simulation and the real world.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-23 in cs.CV).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.25627 — 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 2026-10-08T15:56:18.237Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).