Recent image classification models must balance local feature modeling, cross-window interaction, and parameter efficiency. Many high-performing architectures rely on fully trainable token-mixers, which improve representation learning but increase parameter count, optimization complexity and computational cost. We propose a parameter-efficient image classification model called HiRo that integrates shifted-window partitioning with multi-directional hierarchical reservoir computing. Images are divided into non-overlapping patches (treated as tokens), linearly projected, normalized, and enriched with 2D sinusoidal positional encodings, then processed within local windows. Inside each window, tokens are scanned in four directions and passed through a two-stage slice-and-mix reservoir module. In the first stage, directional sequences are split into contiguous slices, each processed by its own fixed reservoir with a trainable closed-loop readout. The resulting slice outputs are summarized using the start, end, and mean representations, and then mixed by a second-stage fixed reservoir for each direction. The mixed slice representations are expanded back to the token level and fused with the first-stage outputs, after which the four directional outputs are realigned and averaged. Consecutive blocks alternate between regular and shifted windows to enable cross-window interaction, followed by layer normalization, a residual feed-forward network, and global pooling for classification. This design combines regular and shifted window partitioning with hierarchical multi-directional reservoirs to make an efficient local-to-cross-window token-mixing framework for image classification. Despite using under 1M trainable parameters and significantly lower memory and time than transformer-style baselines, HiRo also achieves 99.46%, 85.57%, and 59.10% accuracy on MNIST, CIFAR-10, and CIFAR-100, respectively.

翻译：近期图像分类模型需兼顾局部特征建模、跨窗口交互与参数效率。高性能架构多依赖全可训练令牌混合器，虽能提升表征学习能力，却增加了参数量、优化复杂度与计算成本。我们提出参数高效图像分类模型HiRo，将滑动窗口分区与多方向分层储层计算相融合。图像被划分为非重叠图块（视为令牌），经线性投影、归一化与二维正弦位置编码增强后，于局部窗口内处理。窗口内令牌沿四个方向扫描，通过两阶段切片混合储层模块：第一阶段将方向序列分割为连续切片，每切片由独立固定储层配合可训练闭环读出层处理，以首尾均值表示对切片输出进行摘要；第二阶段经固定储层对各方向进行混合，再将混合切片表示扩展回令牌层级并与第一阶段输出融合，最终将四方向输出重对齐并平均。连续模块交替使用常规窗口与滑动窗口实现跨窗口交互，随后接层归一化、残差前馈网络及全局池化分类。该设计通过组合常规/滑动窗口分区与分层多方向储层，构建了从局部到跨窗口的高效令牌混合框架。尽管可训练参数不足百万，且内存占用与运算时间显著低于Transformer类基线模型，HiRo在MNIST、CIFAR-10与CIFAR-100上仍分别达到99.46%、85.57%与59.10%的准确率。