Search papers, labs, and topics across Lattice.
Meituan
4
0
7
6
OPSD-V shows that incorporating real video context during training can dramatically enhance the performance of autoregressive video generators, leading to superior visual quality and motion fidelity.
Open-source LongCat-Video-Avatar 1.5 leapfrogs closed-source competitors in audio-driven video generation by prioritizing practical engineering over architectural novelty, delivering commercial-grade quality and speed.
Generate realistic talking avatars 120x faster by distilling multi-step diffusion models into a single-step generator, without sacrificing quality.
Generate ultra-high-resolution videos with photorealism and fidelity using a novel latent-cascaded approach that cleverly balances computational cost and detail.
