Bridging the Discrete-Continuous Gap: Unified Multimodal Generation via Coupled Manifold Discrete Absorbing Diffusion

TL;DR

This paper introduces CoM-DAD, a unified multimodal generative framework that combines discrete and continuous diffusion processes to improve stability and alignment in text-image generation.

Contribution

The paper proposes a novel hierarchical dual-process framework that decouples semantic planning from token synthesis, enabling stable, scalable multimodal generation.

Findings

Outperforms existing models in stability and quality

Effectively aligns modalities without heavy contrastive encoders

Establishes a new paradigm for unified text-image generation

Abstract

The bifurcation of generative modeling into autoregressive approaches for discrete data (text) and diffusion approaches for continuous data (images) hinders the development of truly unified multimodal systems. While Masked Language Models (MLMs) offer efficient bidirectional context, they traditionally lack the generative fidelity of autoregressive models and the semantic continuity of diffusion models. Furthermore, extending masked generation to multimodal settings introduces severe alignment challenges and training instability. In this work, we propose \textbf{CoM-DAD} (\textbf{Co}upled \textbf{M}anifold \textbf{D}iscrete \textbf{A}bsorbing \textbf{D}iffusion), a novel probabilistic framework that reformulates multimodal generation as a hierarchical dual-process. CoM-DAD decouples high-level semantic planning from low-level token synthesis. First, we model the semantic manifold via a continuous latent diffusion process; second, we treat token generation as a discrete absorbing diffusion process, regulated by a \textbf{Variable-Rate Noise Schedule}, conditioned on these evolving semantic priors. Crucially, we introduce a \textbf{Stochastic Mixed-Modal Transport} strategy that aligns disparate modalities without requiring heavy contrastive dual-encoders. Our method demonstrates superior stability over standard masked modeling, establishing a new paradigm for scalable, unified text-image generation.