DSS-GAN is the first generative adversarial network to use a Mamba (State Space Model) backbone for high-quality image synthesis.
March 19, 2026
Original Paper
DSS-GAN: Directional State Space GAN with Mamba backbone for Class-Conditional Image Synthesis
arXiv · 2603.17637
The Takeaway
It demonstrates that hierarchical SSMs can outperform StyleGAN2-ADA in precision and FID while maintaining linear scaling. This marks a significant shift in generative architecture away from purely convolutional or transformer-based backbones.
From the abstract
We present DSS-GAN, the first generative adversarial network to employ Mamba as a hierarchical generator backbone for noise-to-image synthesis. The central contribution is Directional Latent Routing (DLR), a novel conditioning mechanism that decomposes the latent vector into direction-specific subvectors, each jointly projected with a class embedding to produce a feature-wise affine modulation of the corresponding Mamba scan. Unlike conventional class conditioning that injects a global signal, D