Classifier-Free Guidance inside the Attraction Basin May Cause Memorization

• URL: http://arxiv.org/abs/2411.16738v2
• Date: Mon, 17 Mar 2025 15:50:58 GMT
• Title: Classifier-Free Guidance inside the Attraction Basin May Cause Memorization
• Abstract summary: Diffusion models are prone to exactly reproduce images from the training data.<n>Memorized images are high in image quality and well-aligned with the conditioning mechanism.
• Score: 16.752023123940674
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Diffusion models are prone to exactly reproduce images from the training data. This exact reproduction of the training data is concerning as it can lead to copyright infringement and/or leakage of privacy-sensitive information. In this paper, we present a novel perspective on the memorization phenomenon and propose a simple yet effective approach to mitigate it. We argue that memorization occurs because of an attraction basin in the denoising process which steers the diffusion trajectory towards a memorized image. However, this can be mitigated by guiding the diffusion trajectory away from the attraction basin by not applying classifier-free guidance until an ideal transition point occurs from which classifier-free guidance is applied. This leads to the generation of non-memorized images that are high in image quality and well-aligned with the conditioning mechanism. To further improve on this, we present a new guidance technique, opposite guidance, that escapes the attraction basin sooner in the denoising process. We demonstrate the existence of attraction basins in various scenarios in which memorization occurs, and we show that our proposed approach successfully mitigates memorization.

Related papers

• Your Pre-trained Diffusion Model Secretly Knows Restoration [55.7186754179308]
  We show that pre-trained diffusion models inherently possess restoration behavior, which can be unlocked by directly learning prompt embeddings.<n>We introduce our lightweight learned prompts on the pre-trained WAN video model and FLUX image models, converting them into high-performing restoration models.
  arXiv  Detail & Related papers  (2026-04-06T17:59:04Z)
• You Don't Need All That Attention: Surgical Memorization Mitigation in Text-to-Image Diffusion Models [8.429432661292964]
  Generative models have been shown to "memorize" certain training data, leading to verbatim or near-verbatim generating images.<n>We introduce Guidance Using Attractive-Repulsive Dynamics (GUARD), a novel framework for memorization mitigation in text-to-image diffusion models.<n>GUARD adjusts the image denoising process to guide the generation away from an original training image and towards one that is distinct from training data.
  arXiv  Detail & Related papers  (2026-02-23T17:20:40Z)
• CAPTAIN: Semantic Feature Injection for Memorization Mitigation in Text-to-Image Diffusion Models [60.610268549138375]
  Diffusion models can unintentionally reproduce training examples, raising privacy and copyright concerns.<n>We introduce CAPTAIN, a training-free framework that mitigates memorization by directly modifying latent features during denoising.
  arXiv  Detail & Related papers  (2025-12-11T14:01:47Z)
• Adjusting Initial Noise to Mitigate Memorization in Text-to-Image Diffusion Models [10.935602641612888]
  We show that the initial noise sample plays a crucial role in determining when this escape occurs.<n>We propose two mitigation strategies that adjust the initial noise-either collectively or individually-to find and utilize initial samples that encourage earlier basin escape.
  arXiv  Detail & Related papers  (2025-10-08T10:37:29Z)
• How Diffusion Models Memorize [26.711679643772623]
  diffusion models can memorize training data, raising serious privacy and copyright concerns.<n>We show memorization is driven by the overestimation of training samples during early denoising.
  arXiv  Detail & Related papers  (2025-09-30T03:03:27Z)
• Improved Immiscible Diffusion: Accelerate Diffusion Training by Reducing Its Miscibility [62.272571285823595]
  We show how immiscibility eases denoising and improves efficiency.<n>We propose a family of implementations including K-nearest neighbor (KNN) noise selection and image scaling to reduce miscibility.<n>This work establishes a potentially new direction for future research into high-efficiency diffusion training.
  arXiv  Detail & Related papers  (2025-05-24T05:38:35Z)
• Learning Flow Fields in Attention for Controllable Person Image Generation [59.10843756343987]
  Controllable person image generation aims to generate a person image conditioned on reference images. We propose learning flow fields in attention (Leffa), which explicitly guides the target query to attend to the correct reference key. Leffa achieves state-of-the-art performance in controlling appearance (virtual try-on) and pose (pose transfer), significantly reducing fine-grained detail distortion.
  arXiv  Detail & Related papers  (2024-12-11T15:51:14Z)
• Exploring Local Memorization in Diffusion Models via Bright Ending Attention [62.979954692036685]
  "bright ending" (BE) anomaly in text-to-image diffusion models prone to memorizing training images. We propose a simple yet effective method to integrate BE into existing frameworks.
  arXiv  Detail & Related papers  (2024-10-29T02:16:01Z)
• Classifier Guidance Enhances Diffusion-based Adversarial Purification by Preserving Predictive Information [75.36597470578724]
  Adversarial purification is one of the promising approaches to defend neural networks against adversarial attacks. We propose gUided Purification (COUP) algorithm, which purifies while keeping away from the classifier decision boundary. Experimental results show that COUP can achieve better adversarial robustness under strong attack methods.
  arXiv  Detail & Related papers  (2024-08-12T02:48:00Z)
• Iterative Ensemble Training with Anti-Gradient Control for Mitigating Memorization in Diffusion Models [20.550324116099357]
  Diffusion models are known for their tremendous ability to generate novel and high-quality samples. Recent approaches for memory mitigation either only focused on the text modality problem in cross-modal generation tasks or utilized data augmentation strategies. We propose a novel training framework for diffusion models from the perspective of visual modality, which is more generic and fundamental for mitigating memorization.
  arXiv  Detail & Related papers  (2024-07-22T02:19:30Z)
• Rethinking and Defending Protective Perturbation in Personalized Diffusion Models [21.30373461975769]
  We study the fine-tuning process of personalized diffusion models (PDMs) through the lens of shortcut learning. PDMs are susceptible to minor adversarial perturbations, leading to significant degradation when fine-tuned on corrupted datasets. We propose a systematic defense framework that includes data purification and contrastive decoupling learning.
  arXiv  Detail & Related papers  (2024-06-27T07:14:14Z)
• Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration [64.84134880709625]
  We show that it is possible to perform domain adaptation via the noise space using diffusion models. In particular, by leveraging the unique property of how auxiliary conditional inputs influence the multi-step denoising process, we derive a meaningful diffusion loss. We present crucial strategies such as channel-shuffling layer and residual-swapping contrastive learning in the diffusion model.
  arXiv  Detail & Related papers  (2024-06-26T17:40:30Z)
• Memorized Images in Diffusion Models share a Subspace that can be Located and Deleted [15.162296378581853]
  Large-scale text-to-image diffusion models excel in generating high-quality images from textual inputs. Concerns arise as research indicates their tendency to memorize and replicate training data. Efforts within the text-to-image community to address memorization explore causes such as data duplication, replicated captions, or trigger tokens.
  arXiv  Detail & Related papers  (2024-06-01T15:47:13Z)
• Unveiling and Mitigating Memorization in Text-to-image Diffusion Models through Cross Attention [62.671435607043875]
  Research indicates that text-to-image diffusion models replicate images from their training data, raising tremendous concerns about potential copyright infringement and privacy risks. We reveal that during memorization, the cross-attention tends to focus disproportionately on the embeddings of specific tokens. We introduce an innovative approach to detect and mitigate memorization in diffusion models.
  arXiv  Detail & Related papers  (2024-03-17T01:27:00Z)
• Learning Discriminative Shrinkage Deep Networks for Image Deconvolution [122.79108159874426]
  We propose an effective non-blind deconvolution approach by learning discriminative shrinkage functions to implicitly model these terms. Experimental results show that the proposed method performs favorably against the state-of-the-art ones in terms of efficiency and accuracy.
  arXiv  Detail & Related papers  (2021-11-27T12:12:57Z)
• Learning to See Through Obstructions with Layered Decomposition [117.77024641706451]
  We present a learning-based approach for removing unwanted obstructions from moving images. Our method leverages motion differences between the background and obstructing elements to recover both layers. We show that the proposed approach learned from synthetically generated data performs well to real images.
  arXiv  Detail & Related papers  (2020-08-11T17:59:31Z)
• Learning to See Through Obstructions [117.77024641706451]
  We present a learning-based approach for removing unwanted obstructions from a short sequence of images captured by a moving camera. Our method leverages the motion differences between the background and the obstructing elements to recover both layers. We show that training on synthetically generated data transfers well to real images.
  arXiv  Detail & Related papers  (2020-04-02T17:59:12Z)

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.