A constrained optimization layer interleaved with flow-matching denoising enables predictive collision avoidance in VLA models, delivering 82.8% collision avoidance and 81.6% task success on SafeLIBERO with gains over single-step baselines.
and Xu, Xiangru and Grizzle, Jessy W
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.RO 2years
2026 2representative citing papers
A hybrid modeling approach unifies planning and control for robots under signal temporal logics via a discrete local feasibility variable and control barrier functions on a transformed disk workspace to handle nonconvex environments without deadlocks.
citing papers explorer
-
Neuro-Symbolic Safety Guidance for Vision-Language-Action Models via Constrained Flow Matching
A constrained optimization layer interleaved with flow-matching denoising enables predictive collision avoidance in VLA models, delivering 82.8% collision avoidance and 81.6% task success on SafeLIBERO with gains over single-step baselines.
-
Feasibility-aware Hybrid Control for Motion Planning under Signal Temporal Logics
A hybrid modeling approach unifies planning and control for robots under signal temporal logics via a discrete local feasibility variable and control barrier functions on a transformed disk workspace to handle nonconvex environments without deadlocks.