Search papers, labs, and topics across Lattice.
D visual features, we introduce a plug-and-play Cross-View
2
0
3
10
Diffusion models can achieve a 10x performance boost in real-world autonomous driving when optimized for trajectory representation, loss space, and augmented with reinforcement learning.
End-to-end autonomous driving is being reshaped by LLMs and VLMs, but challenges in multimodal fusion and safety remain open.