Exploring Diffusion and Flow Matching Under Generator Matching
This work offers a theoretical perspective for researchers in generative modeling, but it is incremental as it builds on existing paradigms without introducing new empirical results.
The paper tackled the problem of understanding diffusion and flow matching by unifying them under the Generator Matching framework, providing theoretical insights into flow matching's robustness and enabling new model classes.
In this paper, we present a comprehensive theoretical comparison of diffusion and flow matching under the Generator Matching framework. Despite their apparent differences, both diffusion and flow matching can be viewed under the unified framework of Generator Matching. By recasting both diffusion and flow matching under the same generative Markov framework, we provide theoretical insights into why flow matching models can be more robust empirically and how novel model classes can be constructed by mixing deterministic and stochastic components. Our analysis offers a fresh perspective on the relationships between state-of-the-art generative modeling paradigms.