RS-SSM：在状态空间模型中细化被遗忘的细节以用于视频语义分割
RS-SSM: Refining Forgotten Specifics in State Space Model for Video Semantic Segmentation
把 Mamba 类状态空间模型中固定尺寸状态导致的「遗忘」显式建模为可逆的遗忘门矩阵，并用通道级幅度感知做对齐，这一「反演遗忘门补全细节」的思路可直接迁移到视频分割、视频目标跟踪等需要长时序像素级一致性的任务。
原文依据 a Forgetting Gate Information Refiner (FGIR) is proposed to adaptively invert and refine the forgetting gate matrix
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近来，状态空间模型通过线性复杂度的状态空间压缩，展现出高效的视频分割能力。然而，视频语义分割（VSS）需要像素级的时空建模能力，以在语义对象的分割中保持时序一致性。尽管状态空间模型在状态空间压缩过程中能够保留共性语义信息，但固定尺寸的状态空间不可避免地会遗忘特定信息，这限制了模型的像素级分割能力。为解决上述问题，我们提出了一种用于视频语义分割的细化特定信息状态空间模型方法（RS-SSM），对被遗忘的时空特定信息进行互补式细化。具体而言，设计了通道级幅度感知机（CwAP），用于提取并对齐状态空间中特定信息的分布特征。此外，提出了遗忘门信息细化器（FGIR），基于特定信息分布自适应地反演并细化状态空间模型中的遗忘门矩阵。由此，RS-SSM 利用反演后的遗忘门，对状态空间压缩过程中被遗忘的特定信息进行互补式细化，从而增强模型的时空像素级分割能力。在四个 VSS 基准上的大量实验表明，RS-SSM 在保持高计算效率的同时达到了最优性能。代码已发布于 https://github.com/zhoujiahuan1991/CVPR2026-RS-SSM。
英文原文摘要 1580 字符
Recently, state space models have demonstrated efficient video segmentation through linear-complexity state space compression. However, Video Semantic Segmentation (VSS) requires pixel-level spatiotemporal modeling capabilities to maintain temporal consistency in segmentation of semantic objects. While state space models can preserve common semantic information during state space compression, the fixed-size state space inevitably forgets specific information, which limits the models' capability for pixel-level segmentation. To tackle the above issue, we proposed a Refining Specifics State Space Model approach (RS-SSM) for video semantic segmentation, which performs complementary refining of forgotten spatiotemporal specifics. Specifically, a Channel-wise Amplitude Perceptron (CwAP) is designed to extract and align the distribution characteristics of specific information in the state space. Besides, a Forgetting Gate Information Refiner (FGIR) is proposed to adaptively invert and refine the forgetting gate matrix in the state space model based on the specific information distribution. Consequently, our RS-SSM leverages the inverted forgetting gate to complementarily refine the specific information forgotten during state space compression, thereby enhancing the model's capability for spatiotemporal pixel-level segmentation. Extensive experiments on four VSS benchmarks demonstrate that our RS-SSM achieves state-of-the-art performance while maintaining high computational efficiency. The code is available at https://github.com/zhoujiahuan1991/CVPR2026-RS-SSM.