摘要：随着图像生成模型扩展至越来越高的分辨率，全局一致性、局部细节与纹理保真度成为生成质量的关键维度。然而，标准 Flow Matching 对所有空间频率一视同仁，忽略了自然存在的频率层级：高频频带会比粗粒度结构更早变得无法与纯噪声区分。我们提出 WaiT，一种小波感知图像 Transformer，它通过无损小波将生成过程分解为粗粒度和细粒度频带。正如其名称所示，高频频带会等待信号：在粗粒度结构形成之前保持为纯噪声，随后加入流过程进行联合细化。由于标准 FID 会通过激进下采样丢弃细粒度细节，我们提出了更严格的三轴评估协议，以在原生分辨率下衡量质量。在 ImageNet 512×512 上，WaiT 实现了 1.43 的像素空间 FID，并在全部三个维度上达到帕累托最优，同时将采样计算量最多降低 50%。借助我们最大的 20 亿参数模型，我们在 ImageNet 512 分辨率上为像素空间模型创下了 1.3 FID 的新 SOTA。我们的公式化方法即使在纹理保真度上也优于最强的潜空间模型，并且能够无缝扩展至高分辨率 OpenImages 和视频生成；在无需任何算法修改的情况下，其在 Kinetics-600 上实现了 0.84 的 SOTA FVD。

Abstract:As image generation models scale to ever higher resolutions, global coherence, local detail, and texture fidelity become critical axes for generation quality. However, standard flow matching treats all spatial frequencies uniformly, ignoring the natural frequency hierarchy where high-frequency bands become indistinguishable from pure noise far earlier than coarse structures. We introduce WaiT, a Wavelet-aware image Transformer that decomposes generation into coarse and fine bands via lossless wavelets. True to its name, the high-frequency bands wait for the signal: staying pure noise until coarse structure has emerged, then joining the flow for joint refinement. Since standard FID discards fine-grained detail through aggressive downsampling, we introduce a more stringent three-axis evaluation protocol to assess quality at native resolution. On ImageNet 512x512, WaiT achieves a pixel-space FID of 1.43 and is Pareto-optimal across all three axes, reducing sampling compute by up to 50%. With our largest 2B model, we set a new state-of-the-art FID of 1.3 for pixel-space models on ImageNet 512 resolution. Our formulation outperforms even the strongest latent-space models on texture fidelity, and scales seamlessly to high-resolution OpenImages and to video generation, achieving a state-of-the-art FVD of 0.84 on Kinetics-600 with no algorithmic modifications.