Arcee：基于Mamba SSMs的生成式视觉建模可微分循环状态链
Arcee: Differentiable Recurrent State Chain for Generative Vision Modeling with Mamba SSMs
AI总结：
针对Mamba SSMs在视觉建模中跨块丢弃终端状态空间表示的问题，提出Arcee跨块循环状态链，复用前块终端SSR作为后块初始条件，在CelebA-HQ上使FID降低5.4倍，兼容现有变体且无额外参数。
摘要：
状态空间模型（SSMs），尤其是Mamba，正越来越多地被用于长上下文序列建模，通过输入依赖的因果选择性扫描操作提供线性时间聚合。沿着这一方向，最近的“Mamba-for-vision”变体主要探索多种扫描顺序，以放松非序列信号（如图像）的严格因果性。Mamba中选择性扫描操作的传统公式在每个块的状态空间动力学中从零重新初始化，丢弃了前一个块的终端状态空间表示（SSR），而不是保留跨块记忆。Arcee是一种跨块循环状态链，将每个块的终端SSR重用作下一个块的初始条件。块间切换被构建为可微分边界映射，其雅可比矩阵支持跨终端边界的端到端梯度流。关键的实用性在于，Arcee与所有先前的“vision-mamba”变体兼容、无参数且成本恒定可忽略。从建模角度来看，我们将终端SSR视为通过对输入的因果传递诱导的温和方向先验，而非非序列信号本身的估计器。为量化影响，在CelebA-HQ（256×256）上使用Flow Matching进行无条件生成时，Arcee将单扫描顺序Zigzag Mamba基线的FID↓从82.81降低到15.33（降低5.4倍）。高效的CUDA内核和训练代码将被发布以支持严谨且可复现的研究。
英文摘要：
State-space models (SSMs), Mamba in particular, are increasingly adopted for long-context sequence modeling, providing linear-time aggregation via an input-dependent, causal selective-scan operation. Along this line, recent "Mamba-for-vision" variants largely explore multiple scan orders to relax strict causality for non-sequential signals (e.g., images). Rather than preserving cross-block memory, the conventional formulation of the selective-scan operation in Mamba reinitializes each block's state-space dynamics from zero, discarding the terminal state-space representation (SSR) from the previous block. Arcee, a cross-block recurrent state chain, reuses each block's terminal state-space representation as the initial condition for the next block. Handoff across blocks is constructed as a differentiable boundary map whose Jacobian enables end-to-end gradient flow across terminal boundaries. Key to practicality, Arcee is compatible with all prior "vision-mamba" variants, parameter-free, and incurs constant, negligible cost. As a modeling perspective, we view terminal SSR as a mild directional prior induced by a causal pass over the input, rather than an estimator of the non-sequential signal itself. To quantify the impact, for unconditional generation on CelebA-HQ (256$\times$256) with Flow Matching, Arcee reduces FID$\downarrow$ from $82.81$ to $15.33$ ($5.4\times$ lower) on a single scan-order Zigzag Mamba baseline. Efficient CUDA kernels and training code will be released to support rigorous and reproducible research.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
