通过控制尺度参数实现图像与分子生成的可调节输出分辨率。
Flow Along the K-Amplitude for Generative Modeling
- 以尺度参数为时间轴，沿K-幅度进行流匹配生成。
- 在图像和分子生成任务中实现分辨率可控的高质量生成。
- 适合需要精细控制生成细节的应用场景。
本文提出一种新型生成学习范式K-Flow，其沿K-幅度进行流动。其中，k为缩放参数，用于组织频率带（或投影系数），幅度描述这些投影系数的范数。通过引入K-幅度分解，K-Flow将缩放参数作为时间轴实现流匹配。我们从理论基础、能量与时间动力学、实际应用三个方向探讨了K-Flow的六个性质。从实用角度，K-Flow可通过控制不同尺度的信息实现可调节生成。实验验证了其在无条件图像生成、类别条件图像生成及分子组装生成中的有效性。此外，通过三项消融研究，证明了通过调控缩放参数可有效控制图像生成的分辨率。
原文摘要 · Abstract (English)
In this work, we propose a novel generative learning paradigm, K-Flow, an algorithm that flows along the $K$-amplitude. Here, $k$ is a scaling parameter that organizes frequency bands (or projected coefficients), and amplitude describes the norm of such projected coefficients. By incorporating the $K$-amplitude decomposition, K-Flow enables flow matching across the scaling parameter as time. We discuss three venues and six properties of K-Flow, from theoretical foundations, energy and temporal dynamics, and practical applications, respectively. Specifically, from the practical usage perspective, K-Flow allows steerable generation by controlling the information at different scales. To demonstrate the effectiveness of K-Flow, we conduct experiments on unconditional image generation, class-conditional image generation, and molecule assembly generation. Additionally, we conduct three ablation studies to demonstrate how K-Flow steers scaling parameter to effectively control the resolution of image generation.
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。