Social-Mamba：基于状态空间模型的社会感知轨迹预测
Social-Mamba: Socially-Aware Trajectory Forecasting with State-Space Models
AI总结：
本文提出Social-Mamba，通过将社会互动视为结构化序列过程，结合循环Mamba模块和社交三元组分解，实现高效准确的轨迹预测，实验表明其在多个基准上表现优异。
摘要：
人类轨迹预测对于拥挤环境中安全导航至关重要，需要在准确性和计算效率之间取得平衡。高效建模社会互动是密集人群中的关键。然而，大多数最新方法依赖于注意力机制，虽然能捕捉复杂依赖关系，但会带来二次计算成本，随着邻居数量的增加而表现不佳。最近，选择性状态空间模型提供了线性时间的替代方案；然而，其本质上是顺序的，与社会互动的无结构和动态性质不匹配。为此，我们提出了Social-Mamba，一种预测架构，将社会互动重新表述为结构化序列过程。其核心是循环Mamba模块，一个新型模块，能够实现连续的双向信息流。Social-Mamba在以自我为中心的网格上组织代理，并引入社交三元组分解，将互动分解为时间、以自我为中心和目标为中心的扫描。这些通过可学习的社会门和全局扫描动态整合，以生成准确且高效的轨迹预测。在五个轨迹预测基准上的广泛实验表明，Social-Mamba在准确率方面达到最先进的水平，同时提供优越的参数效率和计算可扩展性。此外，将Social-Mamba嵌入到流匹配框架中进一步增强了准确性和效率，使其成为未来轨迹预测研究的灵活且稳健的基础。代码已公开：https://github.com/vita-epfl/Social-Mamba
英文摘要：
Human trajectory forecasting is crucial for safe navigation in crowded environments, requiring models that balance accuracy with computational efficiency. Efficiently modeling social interactions is key to performance in dense crowds. Yet, most recent methods rely on attention mechanisms, which are effective at capturing complex dependencies, but incur quadratic computational costs that scale poorly with the growing number of neighbors. Recently, Selective State-Space Models have provided a linear-time alternative; however, their inherently sequential design is misaligned with the unstructured and dynamic nature of social interactions. To address this challenge, we propose Social-Mamba, a forecasting architecture that reformulates social interactions as structured sequential processes. At its core is the Cycle Mamba block, a novel module that enables continuous bidirectional information flow. Social-Mamba organizes agents on an egocentric grid and introduces social triplet factorization, which decomposes interactions into temporal, egocentric, and goal-centric scans. These are dynamically integrated through a learnable social gate and global scan to generate accurate and efficient trajectory predictions. Extensive experiments on five trajectory forecasting benchmarks show that Social-Mamba achieves state-of-the-art accuracy while offering superior parameter efficiency and computational scalability. Furthermore, embedding Social-Mamba into a flow-matching framework further enhances both accuracy and efficiency, establishing it as a flexible and robust foundation for future trajectory forecasting research. The code is publicly available: https://github.com/vita-epfl/Social-Mamba
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。
