MixTTE: Multi-Level Mixture-of-Experts for Scalable and Adaptive Travel Time Estimation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Jiang, Wenzhao, Han, Jindong, Han, Ruiqian, Liu, Hao
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914236158967808
author Jiang, Wenzhao
Han, Jindong
Han, Ruiqian
Liu, Hao
author_facet Jiang, Wenzhao
Han, Jindong
Han, Ruiqian
Liu, Hao
contents Accurate Travel Time Estimation (TTE) is critical for ride-hailing platforms, where errors directly impact user experience and operational efficiency. While existing production systems excel at holistic route-level dependency modeling, they struggle to capture city-scale traffic dynamics and long-tail scenarios, leading to unreliable predictions in large urban networks. In this paper, we propose \model, a scalable and adaptive framework that synergistically integrates link-level modeling with industrial route-level TTE systems. Specifically, we propose a spatio-temporal external attention module to capture global traffic dynamic dependencies across million-scale road networks efficiently. Moreover, we construct a stabilized graph mixture-of-experts network to handle heterogeneous traffic patterns while maintaining inference efficiency. Furthermore, an asynchronous incremental learning strategy is tailored to enable real-time and stable adaptation to dynamic traffic distribution shifts. Experiments on real-world datasets validate MixTTE significantly reduces prediction errors compared to seven baselines. MixTTE has been deployed in DiDi, substantially improving the accuracy and stability of the TTE service.
format Preprint
id arxiv_https___arxiv_org_abs_2601_02943
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MixTTE: Multi-Level Mixture-of-Experts for Scalable and Adaptive Travel Time Estimation
Jiang, Wenzhao
Han, Jindong
Han, Ruiqian
Liu, Hao
Machine Learning
Multiagent Systems
Accurate Travel Time Estimation (TTE) is critical for ride-hailing platforms, where errors directly impact user experience and operational efficiency. While existing production systems excel at holistic route-level dependency modeling, they struggle to capture city-scale traffic dynamics and long-tail scenarios, leading to unreliable predictions in large urban networks. In this paper, we propose \model, a scalable and adaptive framework that synergistically integrates link-level modeling with industrial route-level TTE systems. Specifically, we propose a spatio-temporal external attention module to capture global traffic dynamic dependencies across million-scale road networks efficiently. Moreover, we construct a stabilized graph mixture-of-experts network to handle heterogeneous traffic patterns while maintaining inference efficiency. Furthermore, an asynchronous incremental learning strategy is tailored to enable real-time and stable adaptation to dynamic traffic distribution shifts. Experiments on real-world datasets validate MixTTE significantly reduces prediction errors compared to seven baselines. MixTTE has been deployed in DiDi, substantially improving the accuracy and stability of the TTE service.
title MixTTE: Multi-Level Mixture-of-Experts for Scalable and Adaptive Travel Time Estimation
topic Machine Learning
Multiagent Systems
url https://arxiv.org/abs/2601.02943