**摘要**
We propose and analyze a conservative drifting method for one-step generative modeling. The method replaces the original displacement-based drifting velocity by a kernel density estimator (KDE)-gradient velocity, namely the difference of the kernel-smoothed data score and the kernel-smoothed model score. This velocity is a gradient field, addressing the non-conservatism issue identified for genera
👤 作者: Krishnakumar Balasubramanian

---
🔗 **[Finite-Particle Convergence Rates for Conservative and Non-Conservative Drifting Models](https://arxiv.org/abs/2605.22795v1)**

> Finite-Particle Convergence Rates for Conservative and Non-Conservative Drifting Models
🏷️ 来源: ArXiv cs.AI
⏱️ 2026-05-23 08:00