ActionCache accelerates flow-based VLA models up to 34× by retrieving and refining cached intermediate actions from similar past contexts, without retraining.
DreamVLA: A vision-language-action model dreamed with comprehensive world knowledge
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Training-Free Acceleration for Vision-Language-Action Models with Action Caching and Refinement
ActionCache accelerates flow-based VLA models up to 34× by retrieving and refining cached intermediate actions from similar past contexts, without retraining.