Search papers, labs, and topics across Lattice.
3
0
7
Bridging the gap between 3D and 4D point cloud understanding, PointATA unlocks surprisingly strong performance with parameter-efficient transfer learning, even surpassing full fine-tuning.
Ditch round-based staleness metrics in asynchronous federated learning: parameter sensitivity reveals a more nuanced picture, boosting performance by up to 6.37%.
By learning to intelligently "zoom in" on relevant image regions, TikArt significantly boosts MLLM performance on fine-grained visual reasoning tasks.
