GenerativeMPC grounds VLM-RAG outputs into dynamic velocity limits, safety margins, and virtual stiffness/damping for whole-body MPC and impedance control, enabling 60% speed reduction near humans and socially-aware bimanual manipulation.
Falcon: Actively decoupled visuomotor policies for loco- manipulation with foundation-model-based coordination
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.RO 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
GeoHAT reports a 79.3% mean success rate on the ManiSkill-HAB mobile manipulation benchmark (23.7% above the strongest baseline) by using gated geometric token injection and a hybrid whole-body action decoder.
MPVI interleaves model-based motion planning with VLAs via VLM completion checking to achieve 113% higher task progress on BEHAVIOR-1K without extra data.
citing papers explorer
-
GenerativeMPC: VLM-RAG-guided Whole-Body MPC with Virtual Impedance for Bimanual Mobile Manipulation
GenerativeMPC grounds VLM-RAG outputs into dynamic velocity limits, safety margins, and virtual stiffness/damping for whole-body MPC and impedance control, enabling 60% speed reduction near humans and socially-aware bimanual manipulation.
-
GeoHAT: Geometry-Adaptive Hybrid Action Transformer for Mobile Manipulation
GeoHAT reports a 79.3% mean success rate on the ManiSkill-HAB mobile manipulation benchmark (23.7% above the strongest baseline) by using gated geometric token injection and a hybrid whole-body action decoder.
-
Make Your VLA More Robust Without More Data By Interleaving Motion Planning
MPVI interleaves model-based motion planning with VLAs via VLM completion checking to achieve 113% higher task progress on BEHAVIOR-1K without extra data.