Search papers, labs, and topics across Lattice.
School of Artificial Intelligence, Xidian University, China
2
0
4
Distilling vision foundation models for segmentation doesn't have to sacrifice out-of-domain generalization: GKD achieves +10.6% gains over standard KD in foundation-to-local transfer.
Achieve 44-48% higher accuracy in UAV landing on oscillating marine platforms by explicitly modeling wave harmonics and fusing frequency-domain and time-domain prediction.
