Diffusion-LM

Diffusion-LM is a non-autoregressive language model introduced by Li et al. (2022) that applies continuous diffusion processes to generate text. It iteratively denoises a sequence of word vectors, offering controllable generation and improved perform

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.