Search papers, labs, and topics across Lattice.
1
0
3
Turns out, even the best open-source Vision Language Models are still quite deaf when it comes to sign language recognition, but scaling up the model and training data helps it listen.