Graph convolutional network-based multi-threaded mixing Mamba transformer for traffic forecasting

Xingyu Han

doi:10.1016/j.asoc.2026.116049

Graph convolutional network-based multi-threaded mixing Mamba transformer for traffic forecasting

2026-07-21

0

PRE

AI

DOI：10.1016/j.asoc.2026.116049

Original

Original request for help

Save

Abstract

En 中文

• MST is proposed to enable parallel multi-threaded learning of different features. • The integration of Mamba empowers the model with dynamic inference capabilities. • DA-GCN enables real-time spatio-temporal modeling capabilities. • MixMambaFormer is proposed to achieve high-precision traffic prediction.

Graph convolutional network-based multi-threaded mixing Mamba transformer for traffic forecasting

Abstract

Journal

Organization