Generative Modeling via Drifting

• URL: http://arxiv.org/abs/2602.04770v2
• Date: Fri, 06 Feb 2026 07:18:33 GMT
• Title: Generative Modeling via Drifting
• Authors: Mingyang Deng, He Li, Tianhong Li, Yilun Du, Kaiming He,
• Abstract summary: We propose a new paradigm called Drifting Models, which evolve the pushforward distribution during training and naturally admit one-step inference.<n>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.
• Score: 63.351930190408545
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• Temporal Score Rescaling for Temperature Sampling in Diffusion and Flow Models [30.36841165328262]
  We present a mechanism to steer the sampling diversity of flow matching models.<n>We show that rescaling these allows one to effectively control a local' sampling temperature.
  arXiv  Detail & Related papers  (2025-10-01T17:59:51Z)
• Score-based Idempotent Distillation of Diffusion Models [0.9367224590861915]
  Idempotent generative networks (IGNs) are a new line of generative models based on idempotent mapping to a target manifold.<n>In this work, we unite diffusion and IGNs by distilling idempotent models from diffusion model scores, called SIGN.<n>Our proposed method is highly stable and does not require adversarial losses. We provide a theoretical analysis of our proposed score-based training methods and empirically show that IGNs can be effectively distilled from a pre-trained diffusion model.
  arXiv  Detail & Related papers  (2025-09-25T19:36:10Z)
• Model Integrity when Unlearning with T2I Diffusion Models [11.321968363411145]
  We propose approximate Machine Unlearning algorithms to reduce the generation of specific types of images, characterized by samples from a forget distribution'' We then propose unlearning algorithms that demonstrate superior effectiveness in preserving model integrity compared to existing baselines.
  arXiv  Detail & Related papers  (2024-11-04T13:15:28Z)
• Tuning Timestep-Distilled Diffusion Model Using Pairwise Sample Optimization [97.35427957922714]
  We present an algorithm named pairwise sample optimization (PSO), which enables the direct fine-tuning of an arbitrary timestep-distilled diffusion model.<n>PSO introduces additional reference images sampled from the current time-step distilled model, and increases the relative likelihood margin between the training images and reference images.<n>We show that PSO can directly adapt distilled models to human-preferred generation with both offline and online-generated pairwise preference image data.
  arXiv  Detail & Related papers  (2024-10-04T07:05:16Z)
• Constrained Diffusion Models via Dual Training [80.03953599062365]
  Diffusion processes are prone to generating samples that reflect biases in a training dataset. We develop constrained diffusion models by imposing diffusion constraints based on desired distributions. We show that our constrained diffusion models generate new data from a mixture data distribution that achieves the optimal trade-off among objective and constraints.
  arXiv  Detail & Related papers  (2024-08-27T14:25:42Z)
• One-step Diffusion with Distribution Matching Distillation [54.723565605974294]
  We introduce Distribution Matching Distillation (DMD), a procedure to transform a diffusion model into a one-step image generator. We enforce the one-step image generator match the diffusion model at distribution level, by minimizing an approximate KL divergence. Our method outperforms all published few-step diffusion approaches, reaching 2.62 FID on ImageNet 64x64 and 11.49 FID on zero-shot COCO-30k.
  arXiv  Detail & Related papers  (2023-11-30T18:59:20Z)
• Masked Diffusion Models Are Fast Distribution Learners [32.485235866596064]
  Diffusion models are commonly trained to learn all fine-grained visual information from scratch. We show that it suffices to train a strong diffusion model by first pre-training the model to learn some primer distribution. Then the pre-trained model can be fine-tuned for various generation tasks efficiently.
  arXiv  Detail & Related papers  (2023-06-20T08:02:59Z)
• Learning to Jump: Thinning and Thickening Latent Counts for Generative Modeling [69.60713300418467]
  Learning to jump is a general recipe for generative modeling of various types of data. We demonstrate when learning to jump is expected to perform comparably to learning to denoise, and when it is expected to perform better.
  arXiv  Detail & Related papers  (2023-05-28T05:38:28Z)
• On Distillation of Guided Diffusion Models [94.95228078141626]
  We propose an approach to distilling classifier-free guided diffusion models into models that are fast to sample from. For standard diffusion models trained on the pixelspace, our approach is able to generate images visually comparable to that of the original model. For diffusion models trained on the latent-space (e.g., Stable Diffusion), our approach is able to generate high-fidelity images using as few as 1 to 4 denoising steps.
  arXiv  Detail & Related papers  (2022-10-06T18:03:56Z)
• Can Push-forward Generative Models Fit Multimodal Distributions? [3.8615905456206256]
  We show that the Lipschitz constant of generative networks has to be large in order to fit multimodal distributions. We validate our findings on one-dimensional and image datasets and empirically show that generative models consisting of stacked networks with input at each step do not suffer of such limitations.
  arXiv  Detail & Related papers  (2022-06-29T09:03:30Z)

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.