Search papers, labs, and topics across Lattice.
2
0
3
Cross-modal prediction in MJEPA not only boosts performance but also reveals that a unified encoder can outperform traditional modality-specific approaches by leveraging shared information.
PRIMO lets you quantify exactly how much each missing modality hurts your model's predictions, instance by instance.
