Search papers, labs, and topics across Lattice.
3
0
8
12
Squeeze up to 50 distinct image editing effects, plus few-shot generation, into a single LoRA without sacrificing quality.
Forget static imitation learning: LaST-R1 unlocks near-perfect robotic manipulation (99.8% success) by adaptively reasoning about physical dynamics *before* acting, then refining with RL.
Forget expensive real-world robot training: Hi-WM lets humans directly edit a robot's simulated reality, turning world models into powerful, reusable playgrounds for failure recovery.
