We introduce beta diffusion, a novel generative modeling method that
int...
Diffusion-based models have shown the merits of generating high-quality
...
Through prompting, large-scale pre-trained models have become more expre...
Diffusion models are powerful, but they require a lot of time and data t...
Although text-to-image diffusion models have made significant strides in...
Blind face restoration usually synthesizes degraded low-quality data wit...
Learning the distribution of a continuous or categorical response variab...
For stable training of generative adversarial networks (GANs), injecting...
A topic model is often formulated as a generative model that explains ho...
Offline reinforcement learning enables learning from a fixed dataset, wi...
Employing a forward Markov diffusion chain to gradually map the data to ...
Token-mixing multi-layer perceptron (MLP) models have shown competitive
...
The neural attention mechanism has been incorporated into deep neural
ne...
To measure the difference between two probability distributions, we prop...
Surrogate task based methods have recently shown great promise for
unsup...
Graphs with complete node attributes have been widely explored recently....
Leveraging well-established MCMC strategies, we propose MCMC-interactive...
Graph structured data provide two-fold information: graph structures and...
We focus on an important yet challenging problem: using a 2D deep networ...
Integrating multi-phase information is an effective way of boosting visu...
In information theory, Fisher information and Shannon information (entro...
Variational Autoencoder (VAE) is one of the most popular generative mode...