探索用于文本条件图像生成的 GANs 中的 Sparse MoE
Exploring Sparse MoE in GANs for Text-conditioned Image Synthesis
AI总结：
本文提出 Aurora，一种基于 GAN 的文本到图像生成器，通过引入稀疏激活的混合专家（MoE）和稀疏路由器，在有限计算资源下实现了高效的文本条件图像合成，并在 MS COCO 上取得了 6.2 的零样本 FID。
摘要：
由于难以扩大规模，生成对抗网络（GANs）在文本条件图像合成任务上似乎正逐渐失宠。稀疏激活的混合专家（MoE）最近被证明是利用有限计算资源训练大规模模型的有效解决方案。受此理念启发，我们提出了 Aurora，一种基于 GAN 的文本到图像生成器，它采用一组专家来学习特征处理，并配有一个稀疏路由器来帮助为每个特征点选择最合适的专家。为了忠实地将采样随机性和文本条件解码到最终合成中，我们的路由器通过考虑文本集成的全局潜码来自适应地做出决策。在 64x64 图像分辨率下，我们的模型在 LAION2B-en 和 COYO-700M 上训练，在 MS COCO 上实现了 6.2 的零样本 FID。我们发布了代码和检查点，以促进社区的进一步发展。
英文摘要：
Due to the difficulty in scaling up, generative adversarial networks (GANs) seem to be falling from grace on the task of text-conditioned image synthesis. Sparsely-activated mixture-of-experts (MoE) has recently been demonstrated as a valid solution to training large-scale models with limited computational resources. Inspired by such a philosophy, we present Aurora, a GAN-based text-to-image generator that employs a collection of experts to learn feature processing, together with a sparse router to help select the most suitable expert for each feature point. To faithfully decode the sampling stochasticity and the text condition to the final synthesis, our router adaptively makes its decision by taking into account the text-integrated global latent code. At 64x64 image resolution, our model trained on LAION2B-en and COYO-700M achieves 6.2 zero-shot FID on MS COCO. We release the code and checkpoints to facilitate the community for further development.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
