Search papers, labs, and topics across Lattice.
3
0
4
4
Autonomous driving VLAs can achieve low-latency inference and SOTA planning by generating actions before CoT, with CoT solely supervising action generation.
A 4B grounding foundation model achieves state-of-the-art visual grounding, outperforming larger models and boosting downstream robotic and autonomous driving performance by up to 24.8% in OOD settings.
FactorDrive redefines autonomous driving by seamlessly integrating spatial-physical evidence into adaptive reasoning, achieving state-of-the-art planning performance.