统一架构处理一到三维数据，效率高且可跨平台部署。
Nd-BiMamba2: A Unified Bidirectional Architecture for Multi-Dimensional Data Processing
- 基于Mamba2模块，设计双向处理与自适应填充机制。
- 在1D、2D、3D数据上均实现高效推理，支持多平台运行。
- 模块化统一架构，降低开发维护成本，适合实际应用。
深度学习模型通常需要为不同维度的数据（如1D时间序列、2D图像和3D体数据）设计专用架构。现有双向模型主要针对序列数据，难以有效扩展到高维场景。为此，本文提出一种新型多维双向神经网络架构Nd-BiMamba2，可高效处理1D、2D和3D数据。该架构基于Mamba2模块，引入创新的双向处理机制与自适应填充策略，有效捕捉多维数据中的双向信息，同时保持计算效率。不同于需为各维度定制架构的方法，Nd-BiMamba2采用统一模块化设计，简化开发与维护成本。为验证其可移植性与灵活性，我们成功将其导出至ONNX和TorchScript，并在多种硬件平台（如CPU、GPU及移动端）上测试。实验表明，Nd-BiMamba2可在多平台上高效运行，展现出实际应用潜力。代码已开源：https://github.com/Human9000/nd-Mamba2-torch
原文摘要 · Abstract (English)
Deep learning models often require specially designed architectures to process data of different dimensions, such as 1D time series, 2D images, and 3D volumetric data. Existing bidirectional models mainly focus on sequential data, making it difficult to scale effectively to higher dimensions. To address this issue, we propose a novel multi-dimensional bidirectional neural network architecture, named Nd-BiMamba2, which efficiently handles 1D, 2D, and 3D data. Nd-BiMamba2 is based on the Mamba2 module and introduces innovative bidirectional processing mechanisms and adaptive padding strategies to capture bidirectional information in multi-dimensional data while maintaining computational efficiency. Unlike existing methods that require designing specific architectures for different dimensional data, Nd-BiMamba2 adopts a unified architecture with a modular design, simplifying development and maintenance costs. To verify the portability and flexibility of Nd-BiMamba2, we successfully exported it to ONNX and TorchScript and tested it on different hardware platforms (e.g., CPU, GPU, and mobile devices). Experimental results show that Nd-BiMamba2 runs efficiently on multiple platforms, demonstrating its potential in practical applications. The code is open-source: https://github.com/Human9000/nd-Mamba2-torch
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。