Search papers, labs, and topics across Lattice.
3
0
3
Generative world action models can run up to 3× faster in closed-loop control without fine-tuning or task degradation by dynamically repairing and reusing intermediate computational states instead of recomputing them from scratch.
Conversational 3D heads no longer need static weights: turning live dyadic audiovisual streams into an online self-supervised training signal makes avatar motion up to 11.1% more coherent as dialogues unfold.
Full-body humanoid loco-manipulation no longer requires isolated continuous action heads: factoring discrete action tokens across body parts directly within an LLM's native vocabulary outperforms specialized continuous policies when paired with grouped discrete diffusion decoding.
