MORAIDICTIONNAIRE IA
Génératif

Diffusion Model

The generative model that creates images by learning to turn pure noise into signal.

Picture a sculptor starting from a scrambled block of marble — pure noise — and, stroke by stroke, chipping away the dust until a sharp image emerges. That is the core intuition behind a diffusion model: a family of generative models that learn to reverse a gradual corruption process. It is the technology powering DALL·E 2, Stable Diffusion, Midjourney, and Imagen.

Two mirrored processes

A diffusion model rests on two opposing trajectories:

Adding noise at step $t$ can be written:

$$ x_t = \sqrt{\bar{\alpha}_t}\, x_0 + \sqrt{1 - \bar{\alpha}_t}\, \epsilon, \quad \epsilon \sim \mathcal{N}(0, I) $$

The network is trained to predict this noise $\epsilon$, minimizing the error between the true and predicted noise.

Why it dominates image generation

Criterion Diffusion GAN
Training stability High Often unstable
Image diversity Excellent Risk of mode collapse
Generation speed Slow (multi-step) Fast (single pass)
Quality / control Very fine Variable

Text-conditioned generation is achieved by guiding the denoising with an embedding of the caption (via CLIP, for instance), often reinforced by classifier-free guidance.

The genius of diffusion: turning creation into a series of small denoising problems, each far simpler than painting a whole image in one shot.

Explore the full AI dictionary →