Search papers, labs, and topics across Lattice.
Ordos Research Institute, Liaoning Technical University, Inner Mongolia, China
2
0
3
1
MF-VPD reduces model parameters by nearly 77% while boosting performance in visual perception tasks, making it a game-changer for efficient AI applications.
Attention sparsification via the $\alpha$-entmax transform can dramatically enhance CLIP's performance in open-vocabulary tasks by filtering out irrelevant context.
