Search papers, labs, and topics across Lattice.
1
0
3
AVA-Encoder achieves a 73.1% relative improvement in video representation learning, enabling agents to produce cinematic-grade videos with far fewer resources.
