Search papers, labs, and topics across Lattice.
SJTU-DENG-Lab/WLA
1
0
3
WLA models can learn complex tasks from egocentric videos without requiring action annotations, achieving unprecedented success rates in multi-task learning.