Search papers, labs, and topics across Lattice.
4
0
8
SPIN-4DGS achieves unprecedented fidelity in Gaussian splatting for fast motion, outperforming existing methods by a notable margin.
Filtering videos for physical consistency can boost task success rates by over 8%, bridging the simulation-to-reality gap in video generation.
Forget auxiliary encoders and handcrafted losses: LVRPO uses reinforcement learning to directly align language and vision, boosting performance across a range of multimodal tasks.
Stop treating generated images like real ones: GMAIL aligns them as separate modalities in a shared latent space, unlocking significant gains in vision-language tasks.
