Search papers, labs, and topics across Lattice.
Affiliation:
1
0
2
A lightweight model for music AVQA achieves 96% accuracy by leveraging frozen audio encoders, challenging the notion that larger models are always better.
