发表机构
National Taiwan University（国立台湾大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出CNP-Flow框架，通过流反转学习条件源分布，以改进时间流匹配的生成质量，在视频预测、插值和机器人规划任务中表现优异。
AI 中文摘要
我们提出了CNP-Flow，一种用于时间生成任务的流匹配框架，它通过流反转学习条件源分布。标准条件流匹配（FM）通过向量场引入条件信息，并从标准高斯分布中抽取源样本，而CNP-Flow使用条件噪声预测器（CNP）为每个时间条件生成各向同性高斯源。CNP通过流反转获得的源样本进行监督，流反转将观测到的目标通过预训练的FM模型映射回源空间。三阶段流程包括：预训练FM模型、训练CNP、以及使用学习到的源分布对FM模型进行微调，同时保持FM骨干网络架构不变。在视频预测、视频插值和7自由度Franka机器人运动规划任务中，CNP-Flow一致地提升了生成质量。此外，在更少的函数评估次数下，它也能达到基线性能。项目页面：此https URL
英文摘要
We introduce CNP-Flow, a flow matching framework for temporal generation that learns conditional source distributions through flow reversal. Whereas standard conditional flow matching (FM) incorporates conditioning through the vector field and draws source samples from a standard Gaussian, CNP-Flow uses a conditional noise predictor (CNP) to produce an isotropic Gaussian source for each temporal condition. The CNP is supervised by source samples obtained through flow reversal, which maps observed targets backward through a pretrained FM model. A three-stage pipeline pretrains the FM model, trains the CNP, and fine-tunes the FM model using the learned source distribution, while preserving the FM backbone architecture. Across video prediction, video interpolation, and 7-DoF Franka robot motion planning, CNP-Flow consistently improves generation quality. It also matches baseline performance with fewer function evaluations. Project page: https://embodiedai-ntu.github.io/cnpflow
