Metraipolis › archive › dc_aad6d4debd02

FoveaVLA: Foveation Actions Replace the Wrist Camera in VLAs — arXiv:2610.20064

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20064
License (as archived)
CC0-1.0
Archived
2026-10-15T20:02:54.162Z
SHA-256 of the archived text
134e529a1eb5412f9e30b9b1756b9cd1aaf41dc5625b9816da9c42359f307fd0
Stable URL of this page
https://metraipolis.com/documento/dc_aad6d4debd02

How to cite

FoveaVLA: Foveation Actions Replace the Wrist Camera in VLAs — arXiv:2610.20064 — arXiv (en), https://arxiv.org/abs/2610.20064 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_aad6d4debd02 (Metraipolis archive dc_aad6d4debd02, sha256 134e529a1eb5412f9e30b9b1756b9cd1aaf41dc5625b9816da9c42359f307fd0)
BibTeX
@misc{dcaad6d4debd02,
  title = {FoveaVLA: Foveation Actions Replace the Wrist Camera in VLAs — arXiv:2610.20064},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20064},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_aad6d4debd02; sha256 134e529a1eb5412f9e30b9b1756b9cd1aaf41dc5625b9816da9c42359f307fd0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20064. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present FoveaVLA, a Vision-Language-Action (VLA) model that replaces the wrist camera with learned foveation actions. Wrist cameras provide close-up detail but restrict the view to the end-effector's surroundings. External cameras see the whole scene, but the policy receives them at low resolution, so the fine detail needed for manipulation is lost. FoveaVLA recovers this detail by predicting, alongside each action chunk, the interaction region to observe at the next policy query. The policy then receives crops of these regions, taken from the original-resolution images, together with the full views. We build FoveaVLA on $\pi_{0.5}$ in two configurations, Single and Dual, using one or two external stereo cameras with full and foveal RGB and depth inputs, and post-train both on DROID with RoboInter interaction annotations. Across four real-world task suites, FoveaVLA-Single achieves 70.8% success with one external RGB-D view, and FoveaVLA-Dual achieves 77.1%, versus 52.1% for the wrist-equipped $\pi_{0.5}$. Both match the baseline on plain pick-and-place and improve success with distractors, cloth manipulation, and far-away objects.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.20064 — 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 2026-10-15T19:35:48.398Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).