ResVLA anchors generative VLA policies on low-frequency intent predictions and refines high-frequency residuals via diffusion bridges, yielding competitive performance and faster convergence in simulation and real-robot tests.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
From Noise to Intent: Anchoring Generative VLA Policies with Residual Bridges
ResVLA anchors generative VLA policies on low-frequency intent predictions and refines high-frequency residuals via diffusion bridges, yielding competitive performance and faster convergence in simulation and real-robot tests.