Search papers, labs, and topics across Lattice.
4
1
3
22
Sparse graph policies not only outperform traditional image-based methods but also expose hidden dataset biases, enhancing both performance and interpretability in robot learning.
SPARC reduces noisy labels by leveraging task structure, enabling robots to learn from more reliable demonstrations and outperforming traditional methods in real-world applications.
Current dual-arm manipulation policies falter in real-world applications, with significant challenges in early interactions and skill transfer from simulation.
Decoupling modality processing in VLA models leads to a staggering 95.2% success rate in complex manipulation tasks, far surpassing traditional synchronous approaches.
