面向生成建模的三体散射

摘要

现代生成模型通常依赖于对抗性判别器、预定义的噪声到数据路径或自回归分解。相反，我们证明适当的分布能量可以诱导样本级运动，并为一步生成器提供直接回归监督。用于生成的三体散射建模（TBSM）将能量距离转化为恒定大小的每次抛射体交互：每个抛射体被吸引向一个真实源，并被一个独立生成的源排斥。以抛射体及其条件为条件，该期望等于½ D_E²(P_θ, Q)的2-Wasserstein梯度流速度。一批B个冻结目标事件产生O(B)个样本级损失，每个损失使用一个参考作为其条件，而非像漂移模型（Drifting Models）等方法所使用的整个小批次全配对场。在线追踪这一条件期望可减少场噪声。使用冻结图像特征中的散射，TBSM在ImageNet-256上训练一步生成器，在像素空间PixelDiT-XL上实现FID=2.23，在潜在空间DiT-XL上在NFE=1时实现FID=1.63。我们提供了关联扩散相关监督、类漂移动力学和类GAN目标的设计图谱。这些结果确立了追踪散射作为高维一步生成的一条路径。代码：https://github.com/sp12138/TBSM。

English

Modern generative models typically rely on an adversarial critic, a prescribed noise-to-data path, or an autoregressive factorization. Instead, we show that a proper distributional energy can induce sample-level motion and provide direct regression supervision for a one-step generator. Three-Body Scattering Modeling (TBSM) for generation turns the energy distance into a constant-size per-projectile interaction: each projectile is attracted toward one real source and repelled from one independently generated source. Conditioned on the projectile and its condition, its expectation equals the 2-Wasserstein gradient-flow velocity of frac12D_E^2(P_θ,Q). A batch of B frozen-target events yields O(B) sample-level losses, each using one reference for its condition instead of the minibatch-wide all-pairs field used by methods such as Drifting Models. Tracking this conditional expectation online can reduce field noise. Using scattering in frozen image features, TBSM trains one-step generators on ImageNet-256, achieving FID{}=2.23 with pixel-space PixelDiT-XL and FID{}=1.63 with latent-space DiT-XL at NFE{}=1. We provide a design map relating diffusion-related supervision, Drift-like dynamics, and GAN-like objectives. These results establish tracked scattering as a route to high-dimensional one-step generation. Code: https://github.com/sp12138/TBSM.