Search papers, labs, and topics across Lattice.
3
0
6
2
GaussVLA achieves a 19.7% improvement in spatial manipulation success rates while being more parameter-efficient than previous models.
By representing visual inputs as 3D Gaussian primitives, GST-VLA unlocks a new level of geometric understanding for vision-language-action models, leading to substantial performance gains in robotic manipulation tasks.
Unlock clinically interpretable Parkinson's gait analysis by fusing RGB-D data with a frozen LLM, providing both high accuracy and clear visual-linguistic reasoning.
