Search papers, labs, and topics across Lattice.
Wuhan University
2
0
5
LVLMs can run 2.3x faster with only a 2% accuracy drop, thanks to a new pruning method that understands which visual tokens are most relevant to the text.
Achieve superior remote sensing performance at a fraction of the cost by intelligently sampling high-resolution imagery based on global context and fine-grained patch importance.
