<ul data-eligibleForWebStory="false"><li>The article discusses a new framework called Denoising Multi-Beta VAE that aims to balance between disentanglement and generation quality in generative models.</li><li>Traditionally, achieving interpretable latent representations in generative models comes at the expense of generation quality. The $eta$-VAE method introduces a hyperparameter $eta$ to manage the trade-off between disentanglement and reconstruction quality.</li><li>The Denoising Multi-Beta VAE framework aims to address the disentanglement-reconstruction quality trade-off by utilizing a range of $eta$ values to learn multiple corresponding latent representations. It leverages a non-linear diffusion model to transition between latent representations smoothly.</li><li>The proposed framework is evaluated for its disentanglement and generation quality, showing promising results in achieving both sharp reconstructions and consistent manipulation of generated outputs with respect to changes in $eta.</li></ul>

Denoising Multi-Beta VAE: Representation Learning for Disentanglement and Generation

Discover more