Search papers, labs, and topics across Lattice.
7
0
9
23
Achieving high-fidelity 3D generation with just 1.5% of the training data could revolutionize resource allocation in 3D modeling.
TurboVLA achieves 97.7% success in robotic manipulation while using less than 1 GB of VRAM, challenging the dominance of LLM-centric approaches.
Wat3R achieves superior underwater 3D geometry estimation without any annotated data, leveraging unlabeled footage to overcome the challenges of light attenuation and scattering.
HERMES++ achieves state-of-the-art performance in both future point cloud prediction and 3D scene understanding by unifying these tasks within a single driving world model.
Current vision-language-action models falter in dynamic robotic manipulation due to limited data and poor spatiotemporal reasoning, but a new dataset and architecture close the gap.
VideoLLMs can now watch and think *simultaneously*, achieving 15x faster response times and improved accuracy on video understanding tasks.
Unleashing powerful reasoning in OLLMs doesn't require expensive training data or compute – just clever guidance from existing Large Reasoning Models.
