Concept
dLLMs
dLLMs are diffusion large language models: text generators that produce a sequence by iteratively denoising it, rather than predicting one token at a time left to right. Think of image diffusion models like Stable Diffusion, except the thing being denoised is a block of…
The rest of “dLLMs” 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→