Metraipolis › archive › dc_bfb840e2707e

Decoupling Vision, Language, and Action for Efficient Multi-Task Robot Policies — arXiv:2609.18374

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.18374
License (as archived)
CC0-1.0
Archived
2026-10-15T16:55:52.905Z
SHA-256 of the archived text
d23eb00c0928e8dc21be1731821f03d1b489a157056596da30e2812cd6fbeb8a
Stable URL of this page
https://metraipolis.com/documento/dc_bfb840e2707e

How to cite

Decoupling Vision, Language, and Action for Efficient Multi-Task Robot Policies — arXiv:2609.18374 — arXiv (en), https://arxiv.org/abs/2609.18374 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bfb840e2707e (Metraipolis archive dc_bfb840e2707e, sha256 d23eb00c0928e8dc21be1731821f03d1b489a157056596da30e2812cd6fbeb8a)
BibTeX
@misc{dcbfb840e2707e,
  title = {Decoupling Vision, Language, and Action for Efficient Multi-Task Robot Policies — arXiv:2609.18374},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18374},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bfb840e2707e; sha256 d23eb00c0928e8dc21be1731821f03d1b489a157056596da30e2812cd6fbeb8a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18374. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-Language-Action (VLA) policies commonly run Vision-Language Model (VLM) backbones with billions of parameters at every policy inference, which costs latency and energy. We revisit a decoupled alternative for multi-task manipulation: separate vision and language encoders whose representations condition a compact action head. We run a standardized comparison that varies the vision encoder, the language encoder, and the action head while holding the demonstrations, the training-step budget, the tasks, the evaluation protocol, and the measurement platform fixed, against seven VLA baselines. The resulting Decoupled Embodiment Model (DEM) combines a fine-tuned DINOv3 vision encoder, a frozen NeoBERT language encoder, and a MeanFlow head that generates an action chunk in one forward pass. On 18 RoboCasa tasks evaluated with held-out instruction paraphrases and randomized scenes, DEM reaches 55.6\% mean success against 56.9\% for GR00T N1.7 and 54.6\% for $\pi_{0.5}$, and on three real-robot tasks it reaches 66.0\% against 68.0\% for GR00T N1.7. On the same workstation, DEM needs 6.1\,ms per policy forward pass, a maximum throughput of 162.7 policy calls per second, and draws an estimated 2.07\,J of GPU energy per call, eight to seventeen times the throughput and six to fifteen times less energy than these VLM-backbone policies. Within this trained-task regime, DEM sits on the observed success--latency--energy frontier and provides a strong, efficient baseline for language-conditioned robot skills.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-17 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.18374 — 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 2026-10-15T16:13:27.290Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).