GeoBlock: Inferring Block Granularity from Dependency Geometry in Diffusion Language Models

ArXi:2603.26675v1 Announce Type: cross Block diffusion enables efficient parallel refinement in diffusion language models, but its decoding behavior depends critically on block size. Existing block-sizing strategies rely on fixed rules or heuristic signals and do not account for the dependency geometry that determines which tokens can be safely refined together. This motivates a geometry view of diffusion decoding: \emph{regions with strong causal ordering require sequential updates, whereas semantically cohesive regions admit parallel refinement.} We.