Search papers, labs, and topics across Lattice.
University of Science and Technology of China
3
0
5
Achieve over 90% accuracy in attributing generated videos to their source model with as few as 20 samples, all without training or modifying the videos themselves.
Autoregressive video generation gets a 1.8x speed boost and avoids temporal drift by denoising all blocks hierarchically at the same noise level.
Achieve simultaneous robustness, reversibility, and high capacity in image watermarking with a single watermark, outperforming two-stage methods.
