Search papers, labs, and topics across Lattice.
2
0
4
0
Forget painstakingly annotating paired images: VGGT-Segmentor achieves state-of-the-art cross-view segmentation by cleverly pretraining on single images, sidestepping the need for correspondence.
By integrating hierarchical memory and agentic RL, EventMemAgent enables MLLMs to actively perceive and reason about unbounded video streams, outperforming passive processing methods.
