This section covers generative modeling approaches that learn latent structure and synthesize data from learned distributions.
The paper addresses the "Faithfulness vs. Realism" trade-off in image editing. While GANs are fast, they often require task-specific training and struggle to maintain the user's structural intent (…
Paper:https://arxiv.org/abs/2112.10752 HackMD Link - generate picture from a simple distribtion(e.g.Gaussian distribution noise) - This core concept is to simulate the process of thermodynamic diff…
The authors proposed to use a neural network to find an encoder function $q\psi(x)$ to approximate the true posterior $p{\theta^}(z|x)$, and a decoder function to approximate the true generative mo…
