Search papers, labs, and topics across Lattice.
2
0
5
0
VLA models struggle with physical reasoning, but Pri4R's simple trick of predicting 3D point tracks during training boosts performance by up to 40% on manipulation tasks, without adding any inference overhead.
By explicitly aligning attention with external correspondences, CORAL significantly improves detail preservation in virtual try-on, addressing a key limitation of existing Diffusion Transformer methods.
