Search papers, labs, and topics across Lattice.
National Taiwan University Taipei
4
0
5
Masking spatiotemporal regions in 4D point cloud videos leads to more robust representations that outperform traditional reconstruction methods in downstream tasks.
Audio watermarks can now survive neural resynthesis, thanks to a latent space embedding technique that resists semantic compression by modern audio codecs.
Diffusion models aren't just good for generating synthetic data for robust training; their internal representations can be directly leveraged to boost adversarial robustness and disentangle features.
