Accelerate TarFlow Sampling with GS-Jacobi Iteration

• URL: http://arxiv.org/abs/2505.12849v1
• Date: Mon, 19 May 2025 08:35:44 GMT
• Title: Accelerate TarFlow Sampling with GS-Jacobi Iteration
• Authors: Ben Liu, Zhen Qin,
• Abstract summary: We show that through a series of optimization strategies, TarFlow sampling can be greatly accelerated by using the Gauss-Seidel-Jacobi (abbreviated as GS-Jacobi) iteration method.<n> Experiments on four TarFlow models demonstrate that GS-Jacobi sampling can significantly enhance sampling efficiency while maintaining the quality of generated images.
• Score: 10.411098875443043
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Image generation models have achieved widespread applications. As an instance, the TarFlow model combines the transformer architecture with Normalizing Flow models, achieving state-of-the-art results on multiple benchmarks. However, due to the causal form of attention requiring sequential computation, TarFlow's sampling process is extremely slow. In this paper, we demonstrate that through a series of optimization strategies, TarFlow sampling can be greatly accelerated by using the Gauss-Seidel-Jacobi (abbreviated as GS-Jacobi) iteration method. Specifically, we find that blocks in the TarFlow model have varying importance: a small number of blocks play a major role in image generation tasks, while other blocks contribute relatively little; some blocks are sensitive to initial values and prone to numerical overflow, while others are relatively robust. Based on these two characteristics, we propose the Convergence Ranking Metric (CRM) and the Initial Guessing Metric (IGM): CRM is used to identify whether a TarFlow block is "simple" (converges in few iterations) or "tough" (requires more iterations); IGM is used to evaluate whether the initial value of the iteration is good. Experiments on four TarFlow models demonstrate that GS-Jacobi sampling can significantly enhance sampling efficiency while maintaining the quality of generated images (measured by FID), achieving speed-ups of 4.53x in Img128cond, 5.32x in AFHQ, 2.96x in Img64uncond, and 2.51x in Img64cond without degrading FID scores or sample quality. Code and checkpoints are accessible on https://github.com/encoreus/GS-Jacobi_for_TarFlow

Related papers

• Gaussian Mixture Flow Matching Models [51.976452482535954]
  Diffusion models approximate the denoising distribution as a Gaussian and predict its mean, whereas flow matching models re parameterize the Gaussian mean as flow velocity.<n>They underperform in few-step sampling due to discretization error and tend to produce over-saturated colors under classifier-free guidance (CFG)<n>We introduce a novel probabilistic guidance scheme that mitigates the over-saturation issues of CFG and improves image generation quality.
  arXiv  Detail & Related papers  (2025-04-07T17:59:42Z)
• Normalizing Flows are Capable Generative Models [48.31226028595099]
  TarFlow is a simple and scalable architecture that enables highly performant NF models.<n>It is straightforward to train end-to-end, and capable of directly modeling and generating pixels.<n>TarFlow sets new state-of-the-art results on likelihood estimation for images, beating the previous best methods by a large margin.
  arXiv  Detail & Related papers  (2024-12-09T09:28:06Z)
• Fast constrained sampling in pre-trained diffusion models [77.21486516041391]
  We propose an algorithm that enables fast and high-quality generation under arbitrary constraints.<n>During inference, we can interchange between gradient updates computed on the noisy image and updates computed on the final, clean image.<n>Our approach produces results that rival or surpass the state-of-the-art training-free inference approaches.
  arXiv  Detail & Related papers  (2024-10-24T14:52:38Z)
• Accelerating Auto-regressive Text-to-Image Generation with Training-free Speculative Jacobi Decoding [60.188309982690335]
  We propose a training-free probabilistic parallel decoding algorithm, Speculative Jacobi Decoding (SJD)<n>SJD accelerates the inference of auto-regressive text-to-image generation while maintaining the randomness in sampling-based token decoding.<n>Specifically, SJD facilitates the model to predict multiple tokens at each step and accepts tokens based on the probabilistic criterion.
  arXiv  Detail & Related papers  (2024-10-02T16:05:27Z)
• Boundary-aware Decoupled Flow Networks for Realistic Extreme Rescaling [49.215957313126324]
  Recently developed generative methods, including invertible rescaling network (IRN) based and generative adversarial network (GAN) based methods, have demonstrated exceptional performance in image rescaling. However, IRN-based methods tend to produce over-smoothed results, while GAN-based methods easily generate fake details. We propose Boundary-aware Decoupled Flow Networks (BDFlow) to generate realistic and visually pleasing results.
  arXiv  Detail & Related papers  (2024-05-05T14:05:33Z)
• PaddingFlow: Improving Normalizing Flows with Padding-Dimensional Noise [4.762593660623934]
  We propose PaddingFlow, a novel dequantization method, which improves normalizing flows with padding-dimensional noise. We validate our method on the main benchmarks of unconditional density estimation. The results show that PaddingFlow can perform better in all experiments in this paper.
  arXiv  Detail & Related papers  (2024-03-13T03:28:39Z)
• Scaling Sparse Fine-Tuning to Large Language Models [67.59697720719672]
  Large Language Models (LLMs) are difficult to fully fine-tune due to their sheer number of parameters. We propose SpIEL, a novel sparse finetuning method which maintains an array of parameter indices and the deltas of these parameters relative to their pretrained values. We show that SpIEL is superior to popular parameter-efficient fine-tuning methods like LoRA in terms of performance and comparable in terms of run time.
  arXiv  Detail & Related papers  (2024-01-29T18:43:49Z)
• Comparative Study of Coupling and Autoregressive Flows through Robust Statistical Tests [0.0]
  We propose an in-depth comparison of coupling and autoregressive flows, both of the affine and rational quadratic type. We focus on a set of multimodal target distributions increasing dimensionality ranging from 4 to 400. Our results indicate that the A-RQS algorithm stands out both in terms of accuracy and training speed.
  arXiv  Detail & Related papers  (2023-02-23T13:34:01Z)
• FInC Flow: Fast and Invertible $k \ imes k$ Convolutions for Normalizing Flows [2.156373334386171]
  Invertible convolutions have been an essential element for building expressive normalizing flow-based generative models. We propose a $k times k$ convolutional layer and Deep Normalizing Flow architecture.
  arXiv  Detail & Related papers  (2023-01-23T04:31:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.