Generative Modeling via Drifting
This work addresses the need for high-quality one-step generation in generative modeling, offering a novel approach that could impact applications requiring efficient inference.
The paper tackles the problem of generative modeling by proposing a new paradigm called Drifting Models, which evolves the pushforward distribution during training to enable one-step inference, achieving state-of-the-art results on ImageNet at 256x256 resolution with FID scores of 1.54 in latent space and 1.61 in pixel space.
Generative modeling can be formulated as learning a mapping f such that its pushforward distribution matches the data distribution. The pushforward behavior can be carried out iteratively at inference time, for example in diffusion and flow-based models. In this paper, we propose a new paradigm called Drifting Models, which evolve the pushforward distribution during training and naturally admit one-step inference. We introduce a drifting field that governs the sample movement and achieves equilibrium when the distributions match. This leads to a training objective that allows the neural network optimizer to evolve the distribution. In experiments, our one-step generator achieves state-of-the-art results on ImageNet at 256 x 256 resolution, with an FID of 1.54 in latent space and 1.61 in pixel space. We hope that our work opens up new opportunities for high-quality one-step generation.