# Learning Efficient GANs for Image Translation via Differentiable Masks and co-Attention Distillation

arXiv: https://arxiv.org/abs/2011.08382 · HTML: https://mlsanity.com/paper/2011.08382/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Nov 2020.

## Built on by (most cited first, 17 of 17)

- [Efficient Spatially Sparse Inference for Conditional GANs and Diffusion Models](https://mlsanity.com/paper/2211.02048/) (2022)
- [Distilling a Powerful Student Model via Online Knowledge Distillation](https://mlsanity.com/paper/2103.14473/) (2021)
- [Knowledge Condensation Distillation](https://mlsanity.com/paper/2207.05409/) (2022)
- [HyperStyle3D: Text-Guided 3D Portrait Stylization via Hypernetworks](https://mlsanity.com/paper/2304.09463/) (2023)
- [Cut Inner Layers: A Structured Pruning Strategy for Efficient U-Net GANs](https://mlsanity.com/paper/2206.14658/) (2022)
- [Diversity-aware Channel Pruning for StyleGAN Compression](https://mlsanity.com/paper/2403.13548/) (2024)
- [Discriminator-Cooperated Feature Map Distillation for GAN Compression](https://mlsanity.com/paper/2212.14169/) (2022)
- [Compressing Image-to-Image Translation GANs Using Local Density Structures on Their Learned Manifold](https://mlsanity.com/paper/2312.14776/) (2023)
- [G-NeRF: Geometry-enhanced Novel View Synthesis from Single-View Images](https://mlsanity.com/paper/2404.07474/) (2024)
- [Information-Theoretic GAN Compression with Variational Energy-based Model](https://mlsanity.com/paper/2303.16050/) (2023)
- [LAPP: Layer Adaptive Progressive Pruning for Compressing CNNs from Scratch](https://mlsanity.com/paper/2309.14157/) (2023)
- [SnapCap: Efficient Snapshot Compressive Video Captioning](https://mlsanity.com/paper/2401.04903/) (2024)
- [CoroNetGAN: Controlled Pruning of GANs via Hypernetworks](https://mlsanity.com/paper/2403.08261/) (2024)
- [OvSW: Overcoming Silent Weights for Accurate Binary Neural Networks](https://mlsanity.com/paper/2407.05257/) (2024)
- [Exploring Content Relationships for Distilling Efficient GANs](https://mlsanity.com/paper/2212.11091/) (2022)
- [A Unified Compression Framework for Efficient Speech-Driven Talking-Face Generation](https://mlsanity.com/paper/2304.00471/) (2023)
- [SDAKD: Student Discriminator Assisted Knowledge Distillation for Super-Resolution Generative Adversarial Networks](https://mlsanity.com/paper/2510.03870/) (2025)

## Abstract (arXiv)

Generative Adversarial Networks (GANs) have been widely-used in image translation, but their high computation and storage costs impede the deployment on mobile devices. Prevalent methods for CNN compression cannot be directly applied to GANs due to the peculiarties of GAN tasks and the unstable adversarial training. To solve these, in this paper, we introduce a novel GAN compression method, termed DMAD, by proposing a Differentiable Mask and a co-Attention Distillation. The former searches for a light-weight generator architecture in a training-adaptive manner. To overcome channel inconsistency when pruning the residual connections, an adaptive cross-block group sparsity is further incorporated. The latter simultaneously distills informative attention maps from both the generator and discriminator of a pre-trained model to the searched generator, effectively stabilizing the adversarial training of our light-weight model. Experiments show that DMAD can reduce the Multiply Accumulate Operations (MACs) of CycleGAN by 13x and that of Pix2Pix by 4x while retaining a comparable performance against the full model. Our code can be available at https://github.com/SJLeo/DMAD.