Not All Parameters Matter: Masking Diffusion Models for Enhancing Generation Ability

• URL: http://arxiv.org/abs/2505.03097v1
• Date: Tue, 06 May 2025 01:14:20 GMT
• Title: Not All Parameters Matter: Masking Diffusion Models for Enhancing Generation Ability
• Authors: Lei Wang, Senmao Li, Fei Yang, Jianye Wang, Ziheng Zhang, Yuhan Liu, Yaxing Wang, Jian Yang,
• Abstract summary: MaskUNet is a simple yet effective method-termed MaskUNet''- that enhances generation quality with negligible parameter numbers.<n>We offer two fine-tuning strategies: a training-based approach and a training-free approach, including tailored networks and optimization functions.<n>In zero-shot inference on a COCO dataset, MaskUNet achieves the best FID score and further demonstrates its effectiveness in downstream task evaluations.
• Score: 34.888135351211616
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The diffusion models, in early stages focus on constructing basic image structures, while the refined details, including local features and textures, are generated in later stages. Thus the same network layers are forced to learn both structural and textural information simultaneously, significantly differing from the traditional deep learning architectures (e.g., ResNet or GANs) which captures or generates the image semantic information at different layers. This difference inspires us to explore the time-wise diffusion models. We initially investigate the key contributions of the U-Net parameters to the denoising process and identify that properly zeroing out certain parameters (including large parameters) contributes to denoising, substantially improving the generation quality on the fly. Capitalizing on this discovery, we propose a simple yet effective method-termed ``MaskUNet''- that enhances generation quality with negligible parameter numbers. Our method fully leverages timestep- and sample-dependent effective U-Net parameters. To optimize MaskUNet, we offer two fine-tuning strategies: a training-based approach and a training-free approach, including tailored networks and optimization functions. In zero-shot inference on the COCO dataset, MaskUNet achieves the best FID score and further demonstrates its effectiveness in downstream task evaluations. Project page: https://gudaochangsheng.github.io/MaskUnet-Page/

Related papers

• Striving for Faster and Better: A One-Layer Architecture with Auto Re-parameterization for Low-Light Image Enhancement [50.93686436282772]
  We aim to delve into the limits of image enhancers both from visual quality and computational efficiency.<n>By rethinking the task demands, we build an explicit connection, i.e., visual quality and computational efficiency are corresponding to model learning and structure design.<n>Ultimately, this achieves efficient low-light image enhancement using only a single convolutional layer, while maintaining excellent visual quality.
  arXiv  Detail & Related papers  (2025-02-27T08:20:03Z)
• DepthMaster: Taming Diffusion Models for Monocular Depth Estimation [41.81343543266191]
  We propose a single-step diffusion model designed to adapt generative features for the discriminative depth estimation task.<n>We adopt a two-stage training strategy to fully leverage the potential of the two modules.<n>Our model achieves state-of-the-art performance in terms of generalization and detail preservation, outperforming other diffusion-based methods across various datasets.
  arXiv  Detail & Related papers  (2025-01-05T15:18:32Z)
• WTDUN: Wavelet Tree-Structured Sampling and Deep Unfolding Network for Image Compressed Sensing [51.94493817128006]
  We propose a novel wavelet-domain deep unfolding framework named WTDUN, which operates directly on the multi-scale wavelet subbands. Our method utilizes the intrinsic sparsity and multi-scale structure of wavelet coefficients to achieve a tree-structured sampling and reconstruction.
  arXiv  Detail & Related papers  (2024-11-25T12:31:03Z)
• A Two-Stage Progressive Pre-training using Multi-Modal Contrastive Masked Autoencoders [5.069884983892437]
  We propose a new progressive pre-training method for image understanding tasks which leverages RGB-D datasets. In the first stage, we pre-train the model using contrastive learning to learn cross-modal representations. In the second stage, we further pre-train the model using masked autoencoding and denoising/noise prediction. Our approach is scalable, robust and suitable for pre-training RGB-D datasets.
  arXiv  Detail & Related papers  (2024-08-05T05:33:59Z)
• Neural Network Pruning by Gradient Descent [7.427858344638741]
  We introduce a novel and straightforward neural network pruning framework that incorporates the Gumbel-Softmax technique. We demonstrate its exceptional compression capability, maintaining high accuracy on the MNIST dataset with only 0.15% of the original network parameters. We believe our method opens a promising new avenue for deep learning pruning and the creation of interpretable machine learning systems.
  arXiv  Detail & Related papers  (2023-11-21T11:12:03Z)
• Distance Weighted Trans Network for Image Completion [52.318730994423106]
  We propose a new architecture that relies on Distance-based Weighted Transformer (DWT) to better understand the relationships between an image's components. CNNs are used to augment the local texture information of coarse priors. DWT blocks are used to recover certain coarse textures and coherent visual structures.
  arXiv  Detail & Related papers  (2023-10-11T12:46:11Z)
• PRISTA-Net: Deep Iterative Shrinkage Thresholding Network for Coded Diffraction Patterns Phase Retrieval [6.982256124089]
  Phase retrieval is a challenge nonlinear inverse problem in computational imaging and image processing. We have developed PRISTA-Net, a deep unfolding network based on the first-order iterative threshold threshold algorithm (ISTA) All parameters in the proposed PRISTA-Net framework, including the nonlinear transformation, threshold, and step size, are learned-to-end instead of being set.
  arXiv  Detail & Related papers  (2023-09-08T07:37:15Z)
• Deep Multi-Threshold Spiking-UNet for Image Processing [51.88730892920031]
  This paper introduces the novel concept of Spiking-UNet for image processing, which combines the power of Spiking Neural Networks (SNNs) with the U-Net architecture. To achieve an efficient Spiking-UNet, we face two primary challenges: ensuring high-fidelity information propagation through the network via spikes and formulating an effective training strategy. Experimental results show that, on image segmentation and denoising, our Spiking-UNet achieves comparable performance to its non-spiking counterpart.
  arXiv  Detail & Related papers  (2023-07-20T16:00:19Z)
• MOCA: Self-supervised Representation Learning by Predicting Masked Online Codebook Assignments [72.6405488990753]
  Self-supervised learning can be used for mitigating the greedy needs of Vision Transformer networks. We propose a single-stage and standalone method, MOCA, which unifies both desired properties. We achieve new state-of-the-art results on low-shot settings and strong experimental results in various evaluation protocols.
  arXiv  Detail & Related papers  (2023-07-18T15:46:20Z)
• Semantic Image Synthesis via Diffusion Models [174.24523061460704]
  Denoising Diffusion Probabilistic Models (DDPMs) have achieved remarkable success in various image generation tasks.<n>Recent work on semantic image synthesis mainly follows the de facto GAN-based approaches.<n>We propose a novel framework based on DDPM for semantic image synthesis.
  arXiv  Detail & Related papers  (2022-06-30T18:31:51Z)
• SiPRNet: End-to-End Learning for Single-Shot Phase Retrieval [8.820823270160695]
  convolutional neural networks (CNN) have played important roles in various image reconstruction tasks. In this paper, we design a novel CNN structure, named SiPRNet, to recover a signal from a single Fourier intensity measurement. The proposed approach consistently outperforms other CNN-based and traditional optimization-based methods in single-shot maskless phase retrieval.
  arXiv  Detail & Related papers  (2022-05-23T16:24:52Z)

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.