Discrete Diffusion Models: A Unified Framework from Tokenization to Generation
Provides a unifying perspective for researchers working on discrete diffusion models, clarifying relationships between existing methods and highlighting open challenges.
This paper introduces a unified conceptual framework for discrete diffusion models, showing that existing approaches (transition-matrix, masking, score-based) are instantiations of a common design space. It identifies design trade-offs and suggests future research directions.
Discrete denoising diffusion models (DDMs) have recently emerged as a compelling alternative to autoregressive (AR) modeling for discrete data, offering parallel generation and iterative global refinement capabilities. Unlike continuous diffusion, where the state space is fixed, DDMs are fundamentally shaped by how the discrete state space is constructed: the tokenization scheme, the vocabulary topology, and domain-specific structural alphabets. This work introduces a unified conceptual framework that views discrete diffusion models through the construction of the underlying discrete state space. Within this framework, existing formulations, including transition-matrix, masking/absorbing-state, and score/ratio-based approaches, emerge as different instantiations of a common design space. The framework further exposes common design trade-offs across training objectives, inference algorithms, scaling behavior, systems optimization, and evaluation protocols, suggesting several promising directions for future research.