Next-Concept Modeling: Fast, Parallel Coarse-to-Fine Text Generation

Computer Science > Computation and Language arXiv:2610.14162 (cs) [Submitted on 9 Oct 2026] Title:Next-Concept Modeling: Fast, Parallel Coarse-to-Fine Text Generation Authors:Hermann Kumbong, Ishan S. Khare, Iddah Mlauzi, Andrew Woen, Charles Ding, E. Kelly Buchanan, Simran Arora, Azalia Mirhoseini, Christopher Ré View a PDF of the paper titled Next-Concept Modeling: Fast, Parallel Coarse-to-Fine Text Generation, by Hermann Kumbong and 8 other authors View PDF HTML (experimental) Abstract:Diffusion language models (DLMs) offer parallel sampling for faster generation than autoregressive models, an increasingly important advantage as inference becomes a dominant computational workload. However, DLMs degrade quickly with fewer sampling steps, process the full sequence at…