# Scene-Conditioned Relation Routing for urban cellular activity forecasting

arXiv: https://arxiv.org/abs/2609.20209 · HTML: https://mlsanity.com/paper/2609.20209/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 18 references that are in the map.
- First posted on arXiv 25 Jul 2026.

## Builds on (most cited first, 9 of 9)

- [A Time Series is Worth 64 Words: Long-term Forecasting with Transformers](https://mlsanity.com/paper/2211.14730/) (2022)
- [PDFormer: Propagation Delay-Aware Dynamic Long-Range Transformer for Traffic Flow Prediction](https://mlsanity.com/paper/2301.07945/) (2023)
- [TESTAM: A Time-Enhanced Spatio-Temporal Attention Model with Mixture of Experts](https://mlsanity.com/paper/2403.02600/) (2024)
- [HiNet: Novel Multi-Scenario & Multi-Task Learning with Hierarchical Information Extraction](https://mlsanity.com/paper/2303.06095/) (2023)
- [Adaptive Hybrid Spatial-Temporal Graph Neural Network for Cellular Traffic Prediction](https://mlsanity.com/paper/2303.00498/) (2023)
- [UoMo: A Universal Model of Mobile Traffic Forecasting for Wireless Network Optimization](https://mlsanity.com/paper/2410.15322/) (2024)
- [Cellular Traffic Prediction via Deep State Space Models with Attention Mechanism](https://mlsanity.com/paper/2506.15688/) (2025)
- [UrbanMoE: A Sparse Multi-Modal Mixture-of-Experts Framework for Multi-Task Urban Region Profiling](https://mlsanity.com/paper/2601.22746/) (2026)
- [GraFSTNet: Graph-based Frequency SpatioTemporal Network for Cellular Traffic Prediction](https://mlsanity.com/paper/2602.13282/) (2026)

## Abstract (arXiv)

Urban cellular activity forecasting requires jointly modeling heterogeneous spatiotemporal signals, including SMS usage, mobile network traffic, and call activity. Existing methods often separate temporal modeling, spatial relation learning, and multi-signal prediction, relying on fixed graph structures or static multi-task learning schemes, which limits their adaptability to changing urban scenes. We propose SCRR-Net, a scene-conditioned spatial relation routing framework in which urban contextual information jointly controls spatial dependency selection and cross-task knowledge transfer. SCRR-Net includes a context encoder, a spatial graph expert routing module, a temporal Transformer encoder, and a task knowledge routing module. Experiments on the Milano and Trento datasets demonstrate that SCRR-Net consistently outperforms competing methods on SMS, network traffic, and call activity forecasting, while providing interpretable routing behaviors.