Residual Semantic Steering disentangles visual affordances from language semantics in VLA models using Monte Carlo integration and residual decoding to achieve robustness against linguistic perturbations.
Open the top drawer and put the bowl inside
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
-
Stable Language Guidance for Vision-Language-Action Models
Residual Semantic Steering disentangles visual affordances from language semantics in VLA models using Monte Carlo integration and residual decoding to achieve robustness against linguistic perturbations.