3 citations · 3 across the 6 of their papers we have counts for
1 paper · 2 filters
Daria Pugacheva, Andrey Moskalenko, Denis Shepelev +3
Vision Language Action (VLA) models are widely used in Embodied AI, enabling robots to interpret and execute language instructions. However, their robustness to natural language va…