Fig.1

Concept

Diffusion language model

A diffusion language model generates text by iteratively denoising a sequence of masked or noised tokens, rather than predicting one token at a time left to right. Think of it like an image diffusion model, except the state being refined is a grid of discrete token positions…

The rest of “Diffusion language model” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.

Log in to unlock

← Back to the library