用多尺度拉普拉斯残差加速图像生成，效果更好且更省算力。
Laplacian Multi-scale Flow Matching for Generative Modeling
- 将图像分解为拉普拉斯金字塔残差，通过混合注意力网络并行处理各尺度。
- 在CelebA-HQ和ImageNet上生成质量超越基线，1024×1024图像生成更快更高效。
- 适合追求高分辨率生成且关注推理速度与计算开销的研究者。
本文提出拉普拉斯多尺度流匹配（LapFlow），一种通过多尺度表示提升图像生成建模的新框架。方法将图像分解为拉普拉斯金字塔残差，并利用具有因果注意力机制的混合变压器（MoT）架构并行处理不同尺度，无需像以往级联方法那样在尺度间显式去噪。该多尺度结构不仅提升生成质量，还加快采样速度，促进流匹配方法的可扩展性。在CelebA-HQ和ImageNet上的大量实验表明，该方法在生成样本质量上优于单尺度和多尺度流匹配基线，且所需GFLOPs更少、推理更快。模型可有效扩展至1024×1024高分辨率生成，同时保持较低计算开销。
原文摘要 · Abstract (English)
In this paper, we present Laplacian multiscale flow matching (LapFlow), a novel framework that enhances flow matching by leveraging multi-scale representations for image generative modeling. Our approach decomposes images into Laplacian pyramid residuals and processes different scales in parallel through a mixture-of-transformers (MoT) architecture with causal attention mechanisms. Unlike previous cascaded approaches that require explicit renoising between scales, our model generates multi-scale representations in parallel, eliminating the need for bridging processes. The proposed multi-scale architecture not only improves generation quality but also accelerates the sampling process and promotes scaling flow matching methods. Through extensive experimentation on CelebA-HQ and ImageNet, we demonstrate that our method achieves superior sample quality with fewer GFLOPs and faster inference compared to single-scale and multi-scale flow matching baselines. The proposed model scales effectively to high-resolution generation (up to 1024$\times$1024) while maintaining lower computational overhead.
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。