Search papers, labs, and topics across Lattice.
4
7
5
7
This work instantiates Real-Time EXPO-FT, an RL framework for finetuning real-time VLA policies that meets the real-time control requirements of dynamic real-world manipulation, demonstrating rapid, sample-efficient adaptation to challenging real-world dynamics.
Flow Reversal Steering transforms vague human commands into precise robotic actions, achieving up to 95% higher success rates in real-world tasks with minimal training.
World models can now self-improve by identifying their own prediction errors, thanks to a clever decomposition of action-conditioned prediction into easier-to-verify components.
Q-functions and implicit policy extraction are game-changers for batch online RL in robotics, unlocking significant performance gains over imitation-based approaches.
