Metraipolis › archive › dc_478fcf044801

Text-Vision Synergistic Token Caching: A Training-Free Framework for Efficient Vision-Language-Action Inference — arXiv:2609.34319

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.34319
License (as archived)
CC0-1.0
Archived
2026-10-07T23:52:02.726Z
SHA-256 of the archived text
4207d78259582af3e11cdc17b7b88f56ecd900db51fa04e01e28fffe11a1638f
Stable URL of this page
https://metraipolis.com/documento/dc_478fcf044801

How to cite

Text-Vision Synergistic Token Caching: A Training-Free Framework for Efficient Vision-Language-Action Inference — arXiv:2609.34319 — arXiv (en), https://arxiv.org/abs/2609.34319 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_478fcf044801 (Metraipolis archive dc_478fcf044801, sha256 4207d78259582af3e11cdc17b7b88f56ecd900db51fa04e01e28fffe11a1638f)
BibTeX
@misc{dc478fcf044801,
  title = {Text-Vision Synergistic Token Caching: A Training-Free Framework for Efficient Vision-Language-Action Inference — arXiv:2609.34319},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.34319},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_478fcf044801; sha256 4207d78259582af3e11cdc17b7b88f56ecd900db51fa04e01e28fffe11a1638f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34319. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-Language-Action (VLA) models enable generalizable robotic control but remain computationally expensive. Token caching provides a training-free, plug-and-play acceleration alternative. However, existing VLA caching does not fully exploit a key inductive bias of VLA models: text-vision synergy, wherein textual semantics guide the precise visual grounding of task-relevant regions. In particular, existing designs insufficiently account for head-wise reliability in attention aggregation and layer-wise stability in cache reuse. To address this, we propose Text-Vision Synergistic Token Caching (TVCache), a training-free framework for efficient VLA inference. TVCache filters attention heads based on text-vision information focus to improve task-relevant and physically consistent visual grounding. Concurrently, we introduce a reuse-layer selection mechanism guided by text-vision entropy differences to avoid caching unstable representations and improve cache resource allocation. Extensive experiments across four representative VLA models, two simulation benchmarks, and real-world robotic tasks demonstrate the effectiveness and generality of TVCache. At matched token-retention ratios, TVCache consistently improves task success over existing VLA caching with comparable computational cost. On OpenVLA-OFT, it improves average success by up to 14.5 percentage points over VLA-Cache at 12.5% retention while reducing FLOPs by 2.45x relative to full-token inference.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.34319 — 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 2026-10-07T23:29:27.349Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).