Search papers, labs, and topics across Lattice.
4
0
4
A simple reasoning recipe unlocks massive zero-shot gains in robot foundation models, surpassing the benefits of larger models or more data.
Traditional residual corrections can falter under dynamics shifts, but Warp RL reshapes action distributions to achieve superior performance in challenging environments.
Combining learning and geometric optimization, this framework achieves a 60.9% grasp success rate, outperforming traditional methods by a significant margin.
RoboLab exposes critical performance gaps in leading robotic models, revealing that high-fidelity simulations can better assess generalization than traditional benchmarks.
