让多张图片协作去噪，显著提升生成质量。
Group Diffusion: Enhancing Image Generation by Unlocking Cross-Sample Collaboration
- 让多张图像共享注意力机制，联合去噪。
- 在ImageNet上实现32.2%的FID提升。
- 适合关注生成模型协同推理的研究者。
本文探索了扩散模型推理中被忽视的信号。以往方法在推理时独立生成图像，我们提出Group Diffusion，将注意力机制扩展到跨图像共享，而非仅限于单图内部补丁间。这使得图像可在推理时协同去噪，学习图像内与图像间的对应关系。我们观察到明显的规模效应——组越大，跨样本注意力越强，生成质量越高。此外，我们引入一种定性度量，其强度与FID高度相关。基于标准扩散变压器，GroupDiff在ImageNet-256x256上实现最高32.2%的FID改进。本工作揭示了跨样本推理是生成建模中一种有效且未被探索的新机制。
原文摘要 · Abstract (English)
In this work, we explore an untapped signal in diffusion model inference. While all previous methods generate images independently at inference, we instead ask if samples can be generated collaboratively. We propose Group Diffusion, unlocking the attention mechanism to be shared across images, rather than limited to just the patches within an image. This enables images to be jointly denoised at inference time, learning both intra and inter-image correspondence. We observe a clear scaling effect - larger group sizes yield stronger cross-sample attention and better generation quality. Furthermore, we introduce a qualitative measure to capture this behavior and show that its strength closely correlates with FID. Built on standard diffusion transformers, our GroupDiff achieves up to 32.2% FID improvement on ImageNet-256x256. Our work reveals cross-sample inference as an effective, previously unexplored mechanism for generative modeling.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。