Social-Mamba: Socially-Aware Trajectory Forecasting with State-Space Models

TL;DR

Social-Mamba is a novel social trajectory forecasting model that reformulates social interactions as structured sequential processes, achieving state-of-the-art accuracy with high efficiency and scalability.

Contribution

It introduces a new architecture with Cycle Mamba blocks and social triplet factorization, improving social interaction modeling for trajectory prediction.

Findings

Achieves state-of-the-art accuracy on five benchmarks.

Offers superior parameter efficiency and computational scalability.

Embedding into flow-matching frameworks further improves performance.

Abstract

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