English

Variational Flow Models: Flowing in Your Style

Machine Learning 2024-08-06 v4 Artificial Intelligence

Abstract

We propose a systematic training-free method to transform the probability flow of a "linear" stochastic process characterized by the equation X_{t}=a_{t}X_{0}+\sigma_{t}X_{1} into a straight constant-speed (SC) flow, reminiscent of Rectified Flow. This transformation facilitates fast sampling along the original probability flow via the Euler method without training a new model of the SC flow. The flexibility of our approach allows us to extend our transformation to inter-convert two posterior flows of two distinct linear stochastic processes. Moreover, we can easily integrate high-order numerical solvers into the transformed SC flow, further enhancing the sampling accuracy and efficiency. Rigorous theoretical analysis and extensive experimental results substantiate the advantages of our framework. Our code is available at this [https://github.com/clarken92/VFM||link].

Keywords

Cite

@article{arxiv.2402.02977,
  title  = {Variational Flow Models: Flowing in Your Style},
  author = {Kien Do and Duc Kieu and Toan Nguyen and Dang Nguyen and Hung Le and Dung Nguyen and Thin Nguyen},
  journal= {arXiv preprint arXiv:2402.02977},
  year   = {2024}
}

Comments

Our code is available at: https://github.com/clarken92/VFM

R2 v1 2026-06-28T14:38:29.280Z