Search papers, labs, and topics across Lattice.
2
0
3
0
Incorporating perceptual regularization into diffusion models significantly enhances image super-resolution, preserving fine details that traditional methods often overlook.
Seedance 2.0 leapfrogs existing models by unifying multi-modal inputs (text, image, audio, video) into a single architecture for generating high-quality, longer-duration audio-video content.
