SDiT: Spiking Diffusion Model with Transformer

TL;DR

This paper introduces SDiT, a novel spiking neural network-based diffusion model utilizing transformers to generate high-quality images efficiently, providing a new baseline for SNN generative research.

Contribution

It proposes a unified SNN diffusion architecture with transformers replacing U-net, achieving improved image quality and efficiency over existing SNN generative models.

Findings

High-quality image generation on MNIST, Fashion-MNIST, CIFAR-10

Lower computational cost and shorter sampling time

Competitive performance compared to existing SNN models

Abstract

Spiking neural networks (SNNs) have low power consumption and bio-interpretable characteristics, and are considered to have tremendous potential for energy-efficient computing. However, the exploration of SNNs on image generation tasks remains very limited, and a unified and effective structure for SNN-based generative models has yet to be proposed. In this paper, we explore a novel diffusion model architecture within spiking neural networks. We utilize transformer to replace the commonly used U-net structure in mainstream diffusion models. It can generate higher quality images with relatively lower computational cost and shorter sampling time. It aims to provide an empirical baseline for research of generative models based on SNNs. Experiments on MNIST, Fashion-MNIST, and CIFAR-10 datasets demonstrate that our work is highly competitive compared to existing SNN generative models.