Diffuse Everything: Multimodal Diffusion Models on Arbitrary State Spaces

TL;DR

This paper introduces a novel framework for multimodal diffusion models that operate on arbitrary state spaces, allowing native generation of diverse data types without relying on external preprocessing or encoders.

Contribution

The authors propose a new approach for building multimodal diffusion models on arbitrary state spaces with decoupled noise schedules, enabling native, flexible generation across modalities.

Findings

Effective text-image generation demonstrated

Mixed-type tabular data synthesis achieved

Competitive performance with existing methods

Abstract

Diffusion models have demonstrated remarkable performance in generating unimodal data across various tasks, including image, video, and text generation. On the contrary, the joint generation of multimodal data through diffusion models is still in the early stages of exploration. Existing approaches heavily rely on external preprocessing protocols, such as tokenizers and variational autoencoders, to harmonize varied data representations into a unified, unimodal format. This process heavily demands the high accuracy of encoders and decoders, which can be problematic for applications with limited data. To lift this restriction, we propose a novel framework for building multimodal diffusion models on arbitrary state spaces, enabling native generation of coupled data across different modalities. By introducing an innovative decoupled noise schedule for each modality, we enable both unconditional and modality-conditioned generation within a single model simultaneously. We empirically validate our approach for text-image generation and mixed-type tabular data synthesis, demonstrating that it achieves competitive performance.