Search papers, labs, and topics across Lattice.
1
0
2
Vision-language models struggle significantly with interactive navigation tasks in procedurally generated environments, revealing critical gaps in current AI capabilities.
