6 reports · 2013–2023

Generative models

Adversarial, diffusion and latent-variable models for synthesising images and other high-dimensional data.

Fig. 02 / Field notesStructure, recovered from noise
Four stages of denoising: a noisy seven-by-seven image becomes a structured ring. Schematic illustration.
An illustrative denoising sequence, from noise to a coherent image.

Variational Autoencoders

Generative models · 5 min read · Intermediate

Variational Autoencoders learn compressed latent representations with the ELBO objective and reparameterization trick for generative modeling.

vaevariational-inferencegenerative Read

Generative Adversarial Networks

Generative models · 7 min read · Beginner

An AI-generated review of the GAN framework, in which a generator and a discriminator compete to produce realistic data — one of the most influential ideas.

gangenerative-modelsadversarial-training Read

DDIM: Fast Deterministic Diffusion Sampling

Generative models · 5 min read · Advanced

DDIM turns diffusion sampling deterministic and 10-50x faster with strided steps, keeping DDPM quality while enabling latent interpolation and editing.

diffusionddimsampling Read

Denoising Diffusion Probabilistic Models

Generative models · 8 min read · Advanced

An AI-generated review of DDPM, the 2020 paper that turned diffusion models into the leading family of generative models for images.

diffusiongenerative-modelscomputer-vision Read

ControlNet: Conditional Control for Diffusion Models

Generative models · 5 min read · Intermediate

ControlNet adds spatial conditioning like edges and pose to frozen diffusion models via zero convolutions, enabling precise controllable image generation.

controlnetdiffusionconditioning Read