Variational Flow Models: Flowing in Your Style
This work addresses a bottleneck in generative modeling for researchers and practitioners by providing an incremental improvement to flow-based sampling methods.
The paper tackles the problem of inefficient sampling from probability flows in stochastic processes by introducing a training-free method to transform linear stochastic process flows into straight constant-speed flows, enabling faster and more accurate sampling with high-order solvers, achieving up to 10x speedup in experiments.
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}+σ_{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].