CARE: Certifying Acceleration for Vision-Language-Action Inference
arXiv:2610.08917v1 Announce Type: new Abstract: While vision-language-action (VLA) models have advanced rapidly, running them at every control step remains expensive. Prior work accelerates VLA inference using techniques like action chunking and visual-token pruning, typically evaluating based on l…
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-10-08 04:00 · arXiv cs.CL
CARE: Certifying Acceleration for Vision-Language-Action Inference