Search papers, labs, and topics across Lattice.
Carnegie Mellon University
1
0
2
Navigating with fewer than 8 VLM calls per episode, Goal2Pixel redefines efficiency in vision-language navigation tasks.
