Pith. sign in

SlimFlow: Training Smaller One-Step Diffusion Models with Rectified Flow

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

Diffusion models excel in high-quality generation but suffer from slow inference due to iterative sampling. While recent methods have successfully transformed diffusion models into one-step generators, they neglect model size reduction, limiting their applicability in compute-constrained scenarios. This paper aims to develop small, efficient one-step diffusion models based on the powerful rectified flow framework, by exploring joint compression of inference steps and model size. The rectified flow framework trains one-step generative models using two operations, reflow and distillation. Compared with the original framework, squeezing the model size brings two new challenges: (1) the initialization mismatch between large teachers and small students during reflow; (2) the underperformance of naive distillation on small student models. To overcome these issues, we propose Annealing Reflow and Flow-Guided Distillation, which together comprise our SlimFlow framework. With our novel framework, we train a one-step diffusion model with an FID of 5.02 and 15.7M parameters, outperforming the previous state-of-the-art one-step diffusion model (FID=6.47, 19.4M parameters) on CIFAR10. On ImageNet 64$\times$64 and FFHQ 64$\times$64, our method yields small one-step diffusion models that are comparable to larger models, showcasing the effectiveness of our method in creating compact, efficient one-step diffusion models.

fields

cs.LG 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

Efficient Diffusion Models: A Survey

cs.LG · 2025-02-03 · conditional · novelty 2.0

The paper organizes research on efficient diffusion models into a taxonomy spanning algorithms, systems, and frameworks, and provides a curated reference list.

citing papers explorer

Showing 1 of 1 citing paper.

  • Efficient Diffusion Models: A Survey cs.LG · 2025-02-03 · conditional · none · ref 84 · internal anchor

    The paper organizes research on efficient diffusion models into a taxonomy spanning algorithms, systems, and frameworks, and provides a curated reference list.