Metraipolis › archive › dc_cf4e122b0a63

StrataVLA: Hierarchical and Efficient 3D Geometric Grounding for Vision-Language-Action Models — arXiv:2609.26071

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.26071
License (as archived)
CC0-1.0
Archived
2026-10-08T16:43:47.330Z
SHA-256 of the archived text
8658e06eeb5aa20500c5f8ea83a83c8030fb0a8c9837c5bca2d4e71d6d4f11e7
Stable URL of this page
https://metraipolis.com/documento/dc_cf4e122b0a63

How to cite

StrataVLA: Hierarchical and Efficient 3D Geometric Grounding for Vision-Language-Action Models — arXiv:2609.26071 — arXiv (en), https://arxiv.org/abs/2609.26071 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cf4e122b0a63 (Metraipolis archive dc_cf4e122b0a63, sha256 8658e06eeb5aa20500c5f8ea83a83c8030fb0a8c9837c5bca2d4e71d6d4f11e7)
BibTeX
@misc{dccf4e122b0a63,
  title = {StrataVLA: Hierarchical and Efficient 3D Geometric Grounding for Vision-Language-Action Models — arXiv:2609.26071},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.26071},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cf4e122b0a63; sha256 8658e06eeb5aa20500c5f8ea83a83c8030fb0a8c9837c5bca2d4e71d6d4f11e7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.26071. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-Language-Action (VLA) models inherit strong semantic priors from large-scale vision-language pretraining, yet remain limited in robotic manipulation by insufficient 3D spatial awareness. Existing approaches either require explicit depth or point-cloud inputs, compress geometry into training-time supervision, or inject it only at the model input or action expert, leaving the vision-language backbone without persistent access to task-relevant spatial information. We introduce StrataVLA, a plug-and-play framework for hierarchical geometric grounding. A frozen geometry foundation model extracts shared geometric features from RGB observations, while sparse, layer-specific Geometry Adapters allow visual representations at selected backbone depths to retrieve relevant geometric evidence through cross-attention. To make inference-time geometry practical, StrataVLA further combines task-aware routing with an LRU feature cache that exploits temporal redundancy during task manipulation. Experiments on LIBERO, SimplerEnv, and real-world manipulation demonstrate consistent gains over strong VLA baselines. StrataVLA achieves 98.53% average success on LIBERO suites while reducing geometry-model invocations by up to 88%, establishing hierarchical geometry injection as an effective and efficient way to achieve spatially grounded robotic control.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-23 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.26071 — 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 2026-10-08T15:56:53.480Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).