LaMamba-Diff: Linear-Time High-Fidelity Diffusion Models Based on Local Attention and Mamba

• URL: http://arxiv.org/abs/2408.02615v3
• Date: Thu, 19 Sep 2024 16:07:14 GMT
• Title: LaMamba-Diff: Linear-Time High-Fidelity Diffusion Models Based on Local Attention and Mamba
• Authors: Yunxiang Fu, Chaoqi Chen, Yizhou Yu,
• Abstract summary: Local Attentional Mamba blocks capture both global contexts and local details with linear complexity. Our model exhibits exceptional scalability and surpasses the performance of DiT across various model scales on ImageNet at 256x256 resolution. Compared to state-of-the-art diffusion models on ImageNet 256x256 and 512x512, our largest model presents notable advantages, such as a reduction of up to 62% GFLOPs.
• Score: 54.85262314960038
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent Transformer-based diffusion models have shown remarkable performance, largely attributed to the ability of the self-attention mechanism to accurately capture both global and local contexts by computing all-pair interactions among input tokens. However, their quadratic complexity poses significant computational challenges for long-sequence inputs. Conversely, a recent state space model called Mamba offers linear complexity by compressing a filtered global context into a hidden state. Despite its efficiency, compression inevitably leads to information loss of fine-grained local dependencies among tokens, which are crucial for effective visual generative modeling. Motivated by these observations, we introduce Local Attentional Mamba (LaMamba) blocks that combine the strengths of self-attention and Mamba, capturing both global contexts and local details with linear complexity. Leveraging the efficient U-Net architecture, our model exhibits exceptional scalability and surpasses the performance of DiT across various model scales on ImageNet at 256x256 resolution, all while utilizing substantially fewer GFLOPs and a comparable number of parameters. Compared to state-of-the-art diffusion models on ImageNet 256x256 and 512x512, our largest model presents notable advantages, such as a reduction of up to 62% GFLOPs compared to DiT-XL/2, while achieving superior performance with comparable or fewer parameters. Our code is available at https://github.com/yunxiangfu2001/LaMamba-Diff.

Related papers

• MobileMamba: Lightweight Multi-Receptive Visual Mamba Network [51.33486891724516]
  Previous research on lightweight models has primarily focused on CNNs and Transformer-based designs. We propose the MobileMamba framework, which balances efficiency and performance. MobileMamba achieves up to 83.6% on Top-1, surpassing existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2024-11-24T18:01:05Z)
• MambaUIE&SR: Unraveling the Ocean's Secrets with Only 2.8 GFLOPs [1.7648680700685022]
  Underwater Image Enhancement (UIE) techniques aim to address the problem of underwater image degradation due to light absorption and scattering. Recent years, both Convolution Neural Network (CNN)-based and Transformer-based methods have been widely explored. MambaUIE is able to efficiently synthesize global and local information and maintains a very small number of parameters with high accuracy.
  arXiv  Detail & Related papers  (2024-04-22T05:12:11Z)
• EfficientVMamba: Atrous Selective Scan for Light Weight Visual Mamba [19.062950348441426]
  This work proposes to explore the potential of visual state space models in light-weight model design and introduce a novel efficient model variant dubbed EfficientVMamba. Our EfficientVMamba integrates a atrous-based selective scan approach by efficient skip sampling, constituting building blocks designed to harness both global and local representational features. Experimental results show that, EfficientVMamba scales down the computational complexity while yields competitive results across a variety of vision tasks.
  arXiv  Detail & Related papers  (2024-03-15T02:48:47Z)
• MiM-ISTD: Mamba-in-Mamba for Efficient Infrared Small Target Detection [72.46396769642787]
  We develop a nested structure, Mamba-in-Mamba (MiM-ISTD), for efficient infrared small target detection. MiM-ISTD is $8 times$ faster than the SOTA method and reduces GPU memory usage by 62.2$%$ when testing on $2048 times 2048$ images.
  arXiv  Detail & Related papers  (2024-03-04T15:57:29Z)
• PointMamba: A Simple State Space Model for Point Cloud Analysis [65.59944745840866]
  We propose PointMamba, transferring the success of Mamba, a recent representative state space model (SSM), from NLP to point cloud analysis tasks. Unlike traditional Transformers, PointMamba employs a linear complexity algorithm, presenting global modeling capacity while significantly reducing computational costs.
  arXiv  Detail & Related papers  (2024-02-16T14:56:13Z)
• SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentation [16.476244833079182]
  We introduce SegMamba, a novel 3D medical image textbfSegmentation textbfMamba model. SegMamba excels in whole volume feature modeling from a state space model standpoint. Experiments on the BraTS2023 dataset demonstrate the effectiveness and efficiency of our SegMamba.
  arXiv  Detail & Related papers  (2024-01-24T16:17:23Z)
• MLIC++: Linear Complexity Multi-Reference Entropy Modeling for Learned Image Compression [30.71965784982577]
  We introduce MEM++, which captures diverse range of correlations inherent in the latent representation. MEM++ achieves state-of-the-art performance, reducing BD-rate by 13.39% on the Kodak dataset compared to VTM-17.0 in PSNR. MLIC++ exhibits linear GPU memory consumption with resolution, making it highly suitable for high-resolution image coding.
  arXiv  Detail & Related papers  (2023-07-28T09:11:37Z)
• DilateFormer: Multi-Scale Dilated Transformer for Visual Recognition [62.95223898214866]
  We explore effective Vision Transformers to pursue a preferable trade-off between the computational complexity and size of the attended receptive field. With a pyramid architecture, we construct a Multi-Scale Dilated Transformer (DilateFormer) by stacking MSDA blocks at low-level stages and global multi-head self-attention blocks at high-level stages. Our experiment results show that our DilateFormer achieves state-of-the-art performance on various vision tasks.
  arXiv  Detail & Related papers  (2023-02-03T14:59:31Z)
• Global Filter Networks for Image Classification [90.81352483076323]
  We present a conceptually simple yet computationally efficient architecture that learns long-term spatial dependencies in the frequency domain with log-linear complexity. Our results demonstrate that GFNet can be a very competitive alternative to transformer-style models and CNNs in efficiency, generalization ability and robustness.
  arXiv  Detail & Related papers  (2021-07-01T17:58:16Z)

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.