Diffusion language model – Definition & Meaning

Definition of diffusion language model

: a language model that generates text through an iterative denoising process, repeatedly turning a noisy, masked, or incomplete sequence into a more complete one instead of producing every token strictly from left to right
- The diffusion language model revised several masked positions during each denoising step.
- Researchers compared the DLM with an autoregressive model on controllable generation and parallel decoding.

Origin & history

first known use2022

The 2022 NeurIPS paper Diffusion-LM Improves Controllable Text Generation introduced Diffusion-LM as a non-autoregressive language model based on continuous diffusion. Later work developed discrete and masked diffusion approaches that denoise token sequences directly.

Test yourself

Which of these is the meaning of diffusion language model?

Related terms

diffusion model a generative model that creates images (or audio or video) by starting from random noise and… discrete diffusion a diffusion-model approach for categorical or other discrete data that learns to reverse a… natural language processing the field of AI concerned with getting computers to understand, generate, and work with human… token the basic unit of text a language model reads and writes: a whole word, part of a word, a… constrained decoding a generation technique that restricts which tokens a language model may choose at each… transformer a neural network design that processes a whole sequence at once and uses attention to work out…

Cite this entry

"Diffusion language model." AI Dictionary, Dadgogo, https://dadgogo.com/dictionary/diffusion-language-model/. Accessed 12 Oct. 2026.