Search papers, labs, and topics across Lattice.
1
0
3
Forget brute-force scaling: injecting topological awareness into smaller VLMs drastically improves performance on Vision-Language Navigation, surpassing larger models.
