Search papers, labs, and topics across Lattice.
4
0
4
6
RoboEdit transforms abundant human manipulation videos into high-fidelity training data for robots, unlocking new avenues for scalable robot learning.
LEGO-RL boosts coding agent performance by up to 8.4% while ensuring robust training signals and execution reliability.
Cloning synthetic demonstrations alone caps performance, but combining it with reinforcement learning unlocks superior real-world manipulation capabilities.
Agentic code review can transform AI-generated pull requests into high-quality solutions, outperforming traditional methods in both accuracy and usefulness.
