Search papers, labs, and topics across Lattice.
3
0
8
Robotic world-action models do not need pretrained internet video backbones to scale: training native planning and dynamics modules from scratch across 30,000 hours unlocks strong zero-shot manipulation and drives a 17.7-point gain on an embodiment comprising under 2% of the data.
LLMs, like humans, exhibit a "frequency bias," performing better when prompted and fine-tuned with more common textual expressions.
Image generation models can now achieve state-of-the-art fidelity with up to 64x fewer tokens, thanks to a novel masking strategy that prevents latent space collapse.
