Search papers, labs, and topics across Lattice.
2
0
4
Navigate faster and cheaper: SFCo-Nav cuts VLN token consumption by 50% and speeds up execution by 3.5x by intelligently triggering a slow LLM only when a fast reactive navigator loses confidence.
Achieve state-of-the-art place recognition by fusing camera and LiDAR data with a Visual Geometry Grounded Transformer, and then re-ranking without any additional training.
