提出双分支像素扩散模型，提升图像细节与生成效率
DuSPiT: Dual-Branch Sub-Patch Pixel Diffusion Transformer
- 双分支结构分离全局结构与局部细节建模
- 生成图像细节更丰富，质量-效率平衡更优
- 适合追求高分辨率图像生成的开发者
扩散Transformer在图像生成中表现优异，但多数在压缩隐空间运行。像素空间扩散避免信息损失，但现有方法将每个原始图像块映射为单一标记，迫使单个表示同时处理全局通信与精细细节。为此，我们提出新架构DuSPiT（双分支子块像素扩散Transformer），将全局结构推理与局部外观建模分离。该模型采用紧凑基干分支实现高效全局推理，同时并行使用高容量像素分支，按子块分组以保留精细外观，两分支通过交叉注意力交互。实验表明，DuSPiT生成图像细节更丰富、细粒度结构更强，且在质量-效率权衡上优于先前像素空间扩散Transformer。
原文摘要 · Abstract (English)
Diffusion Transformers achieve strong image generation performance, but most operate in compressed latent spaces. Pixel-space diffusion avoids this information loss, yet existing approaches map each raw image patch to a single token, forcing one representation to handle both global communication and fine-grained details. We address this issue by proposing a new architecture, \textbf{DuSPiT}, a \textbf{Du}al-branch \textbf{S}ub\textbf{P}atch \textbf{Pi}xel \textbf{T}ransformer. This model separates global structural reasoning from local appearance modeling. DuSPiT uses a compact base branch for efficient global reasoning and a parallel, high-capacity pixel branch, organized into subpatch groups, to preserve detailed appearance, with the two branches interacting through cross-attention. Our results show that DuSPiT generates images with richer details and stronger fine-grained structures, while also achieving a better quality--efficiency trade-off than prior pixel-space diffusion transformers.
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。