Multimodal Translation

Diffusion Models

Generative models that learn to progressively denoise data to generate high-quality samples, particularly effective for text-to-image synthesis. These models use forward and reverse diffusion processes for generation.

← Wstecz