Search papers, labs, and topics across Lattice.
1
0
3
MLLMs still struggle to reliably predict the long-term consequences of actions in egocentric videos, even with structured scene annotations.
