← Back to all news
arXiv Machine Learning

A Multi-View Coupled Tensor Decomposition for Lightweight Online Adaptive Traffic Prediction

The paper introduces a Multi-View Coupled Tensor Decomposition (MVCTD) model for online traffic prediction that handles imperfect multi-view data such as speed, flow, and occupancy. MVCTD builds a structured latent forecasting space by jointly modeling shared spatial structures across views and view‑specific temporal dynamics, and incorporates group sparse regularization to mitigate the impact of traffic anomalies. For streaming deployment, the method performs iterative refinement only on the current latent tensor, updating other variables with lightweight closed‑form steps based on summarized historical data, which reduces runtime while maintaining accuracy even under severe missingness.

Related stories

arXiv AI
1d ago

Spatio-Temporal Partial Sensing Forecast for Long-term Traffic

The paper introduces the Spatio-temporal Long-term Partial sensing Forecast model (SLPF) for predicting long-term traffic when sensors are only available at some locations. It tackles challenges such as unknown data distribution at unsensed sites, complex spatio-temporal correlations, and noise by employing a rank-based embedding, a spatial transfer matrix, and a multi-step training process. Experiments on real-world datasets show that SLPF outperforms existing methods.

By Zibo Liu, Zhe Jiang, Zelin Xu, Tingsong Xiao, Zhengkun Xiao, Yupu zhang, Haibo Wang, Shigang Chen
rag
More like this →
arXiv AI
Sep 18

FedeRICo: Federated Region-Influenced Coupling for Traffic Flow Prediction

FedeRICo is a federated traffic forecasting framework that addresses heterogeneity across client sensor subgraphs by combining gradient-level collaboration with boundary-aware residual communication. It uses a dual-branch architecture: a globally guided branch for transferable forecasting structure and a private residual branch that preserves client-specific corrections and incorporates boundary residual signals. Experiments on four real-world traffic benchmarks show that FedeRICo outperforms state‑of‑the‑art federated spatial‑temporal baselines while keeping training runtime competitive.

By Fermin Orozco, Man Luo, Johan Wahlstr\"om
benchmarkssafety
More like this →
arXiv Machine Learning
Sep 11

A Dynamic Fusion Large Language Model for Traffic Flow Prediction

The paper introduces DF-LLM, a Dynamic Fusion Large Language Model designed for traffic flow prediction. It combines a spatiotemporal embedding module, a fusion module that uses graph convolution to capture spatial topology and dynamic dependencies, and an LLM backbone with differentiated parameter adaptation and context aggregation attention. Experiments on four datasets demonstrate that DF-LLM outperforms existing methods in predictive accuracy.

By Xue Qiu, Jianli Xiao
llmsrag
More like this →
arXiv AI
Jul 2

Flow-Through Tensors: A Unified Computational Graph Architecture for Multi-Layer Transportation Network Optimization

arXiv:2507. 02961v2 Announce Type: replace-cross Abstract: Modern transportation network modeling increasingly involves the integration of diverse methodologies including sensor-based forecasting, reinforcement learning, classical flow optimization, and demand modeling that have traditionally been developed in isolation.

By Xuesong Zhou, Taehooie Kim, Mostafa Ameli, Henan Zhu, Yudai Honma, Ram M. Pendyala
reinforcement-learning
More like this →
Hugging Face Trending Papers
Jul 8

Multimodal Spatiotemporal-Frequency Fusion with Peak Enhancement for Cellular Traffic Forecasting

Accurate forecasting of cellular network traffic is essential for network planning, resource allocation, and quality-of-service assurance in modern mobile communication systems. Real-world traffic often exhibits bursty endogenous dynamics and disturbances triggered by external urban events, which makes reliable prediction highly challenging.

ragmultimodal
More like this →