基于时序配对一致性的方差缩减流匹配

摘要

连续时间生成模型（如扩散模型、流匹配和修正流）通过学习时间依赖的向量场进行建模，但传统训练目标通常将不同时间步视为独立处理，导致估计量方差过高和采样效率低下。现有方法通过显式平滑惩罚、轨迹正则化或修改概率路径与求解器来缓解此问题。我们提出时序配对一致性（TPC），这是一种轻量级的降方差原理：通过耦合同一概率路径上配对时间步的速度预测，完全在估计量层面实现优化，无需改变模型架构、概率路径或求解器。理论分析表明，TPC会诱导出二次型的轨迹耦合正则化，在保持流匹配目标不变的同时可证明降低梯度方差。在流匹配框架中实例化TPC后，在CIFAR-10和ImageNet多个分辨率下的实验表明，该方法在相同或更低计算成本下能获得更优的样本质量与效率，其FID指标优于现有方法，并可无缝扩展至现代SOTA风格流程（含噪声增强训练、基于分数的去噪和修正流技术）。

English

Continuous-time generative models, such as diffusion models, flow matching, and rectified flow, learn time-dependent vector fields but are typically trained with objectives that treat timesteps independently, leading to high estimator variance and inefficient sampling. Prior approaches mitigate this via explicit smoothness penalties, trajectory regularization, or modified probability paths and solvers. We introduce Temporal Pair Consistency (TPC), a lightweight variance-reduction principle that couples velocity predictions at paired timesteps along the same probability path, operating entirely at the estimator level without modifying the model architecture, probability path, or solver. We provide a theoretical analysis showing that TPC induces a quadratic, trajectory-coupled regularization that provably reduces gradient variance while preserving the underlying flow-matching objective. Instantiated within flow matching, TPC improves sample quality and efficiency across CIFAR-10 and ImageNet at multiple resolutions, achieving lower FID at identical or lower computational cost than prior methods, and extends seamlessly to modern SOTA-style pipelines with noise-augmented training, score-based denoising, and rectified flow.

基于时序配对一致性的方差缩减流匹配

Temporal Pair Consistency for Variance-Reduced Flow Matching

摘要

Support