Search papers, labs, and topics across Lattice.
3
0
5
R2RDreamer achieves superior spatial generalization for 2D manipulation policies by transforming 3D action-observation pairs into coherent 2D video data, all while minimizing the sim-to-real gap.
By explicitly aligning visuomotor representations with the action space's rigid-body geometry, OASIS achieves superior robotic manipulation performance compared to methods relying on implicit geometry recovery.
Generating robotic manipulation data that respects object affordances is now possible at scale, but current imitation learning methods still struggle with tasks like pouring and hanging, revealing a critical gap.
