Adversarial Learning of Classifier-Free Guidance Schedules

• URL: http://arxiv.org/abs/2608.14038v1
• Date: Fri, 14 Aug 2026 07:30:36 GMT
• Title: Adversarial Learning of Classifier-Free Guidance Schedules
• Abstract summary: We learn the guidance schedule as a function of diffusion time, conditioning and the current noisy sample.<n>We frame this as a density ratio estimation problem.<n>A lightweight generator network predicts the optimal, state-dependent guidance scale.
• Score: 74.70485037604884
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Modern text-to-image diffusion models rely on classifier-free guidance (CFG) to achieve high image fidelity and text alignment. However, CFG typically applies a static, global scale across all timesteps, samples, and conditions -- a choice that is generally suboptimal and can introduce artifacts, as different states may benefit from different levels of guidance. While time-varying schedules are known to improve quality, designing them by hand is non-trivial and application-dependent. In this paper, we learn the guidance schedule as a function of diffusion time, conditioning and the current noisy sample, in order to better align sampled images with the text prompt. We frame this as a density ratio estimation problem: a discriminator is trained to estimate the time-dependent log-density ratio between the true and guided marginal distributions, while a lightweight generator network predicts the optimal, state-dependent guidance scale. Empirically, our approach outperforms both heuristic CFG schedules and prior methods for learning dynamic guidance on text-to-image generation benchmarks.

Related papers

• Information-Theoretic Classifier-Free Guidance with Adaptive Schedule Optimization [15.35957804795103]
  Diffusion models have achieved strong performance in image, text-to-image, and video generation.<n>We propose an information-theoretic framework for CFG schedule optimization.<n>On ImageNet-512 with EDM-XXL and COCO with SD-XL, the learned schedules achieve competitive or improved trade-offs over constant guidance.
  arXiv  Detail & Related papers  (2026-06-23T00:04:04Z)
• HiGFA: Hierarchical Guidance for Fine-grained Data Augmentation with Diffusion Models [82.10385962490051]
  Generative diffusion models show promise for data augmentation.<n>Applying them to fine-grained tasks presents a significant challenge.<n>HiGFA is a hierarchical, confidence-driven orchestration that generates diverse yet faithful synthetic images.
  arXiv  Detail & Related papers  (2025-11-16T10:46:16Z)
• Dynamic Classifier-Free Diffusion Guidance via Online Feedback [53.54876309092376]
  "One-size-all" approach fails to adapt to the diverse requirements of different prompts.<n>We introduce a framework for dynamic CFG scheduling.<n>We demonstrate the effectiveness of our approach on both small-scale models and the state-of-the-art Imagen 3.
  arXiv  Detail & Related papers  (2025-09-19T16:27:19Z)
• Navigating with Annealing Guidance Scale in Diffusion Space [50.53780111249146]
  The choice of the guidance scale has a critical impact on the convergence toward a visually appealing and prompt-adherent image.<n>In this work, we propose an annealing guidance scheduler which dynamically adjusts the guidance scale over time.<n> Empirical results demonstrate that our guidance scheduler significantly enhances image quality and alignment with the text prompt.
  arXiv  Detail & Related papers  (2025-06-30T17:55:00Z)
• Classifier-free Guidance with Adaptive Scaling [7.179513844921256]
  Free guidance (CFG) is an essential mechanism in text-driven diffusion models.<n>In this paper, we present $beta$adaptive-CFG, which controls the impact of guidance during generation.<n>Our model obtained better FID scores, maintaining the text-to-image CLIP similarity scores at a level similar to that of the reference CFG.
  arXiv  Detail & Related papers  (2025-02-14T22:04:53Z)
• Gradient-Free Classifier Guidance for Diffusion Model Sampling [4.450496470631169]
  Gradient-free Guidance (GFCG) method consistently improves class prediction accuracy. For ImageNet 512$times$512, we achieve a record $FD_textDINOv2$ 23.09, while simultaneously attaining a higher classification Precision (94.3%) compared to ATG (90.2%)
  arXiv  Detail & Related papers  (2024-11-23T00:22:21Z)
• Consistency Regularization for Generalizable Source-free Domain Adaptation [62.654883736925456]
  Source-free domain adaptation (SFDA) aims to adapt a well-trained source model to an unlabelled target domain without accessing the source dataset. Existing SFDA methods ONLY assess their adapted models on the target training set, neglecting the data from unseen but identically distributed testing sets. We propose a consistency regularization framework to develop a more generalizable SFDA method.
  arXiv  Detail & Related papers  (2023-08-03T07:45:53Z)
• Imposing Consistency for Optical Flow Estimation [73.53204596544472]
  Imposing consistency through proxy tasks has been shown to enhance data-driven learning. This paper introduces novel and effective consistency strategies for optical flow estimation.
  arXiv  Detail & Related papers  (2022-04-14T22:58:30Z)

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.