Grounded Action Model introduced with a 3D grounding approach to robotics
The announcement proposes building robot foundation models on a pretrained 3D grounding model, while questioning whether language and video models are the right starting point.
TLDR
A post introducing Grounded Action Model (GAM) describes an approach that builds robot foundation models on top of a pretrained 3D grounding model. It questions language and video models as foundations for robotics, summing up the proposed approach as: “Ground first. Then learn to act.”
Grounded Action Model introduced with a 3D grounding approach to robotics
The announcement proposes building robot foundation models on a pretrained 3D grounding model, while questioning whether language and video models are the right starting point.