# Sim-MSTNet: sim2real based Multi-task SpatioTemporal Network Traffic Forecasting

arXiv: https://arxiv.org/abs/2601.21384 · HTML: https://mlsanity.com/paper/2601.21384/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 22 references that are in the map.
- First posted on arXiv 29 Jan 2026.

## Builds on (most cited first, 11 of 11)

- [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting](https://mlsanity.com/paper/2012.07436/) (2020)
- [iTransformer: Inverted Transformers Are Effective for Time Series Forecasting](https://mlsanity.com/paper/2310.06625/) (2023)
- [A Survey of Sim-to-Real Methods in RL: Progress, Prospects and Challenges with Foundation Models](https://mlsanity.com/paper/2502.13187/) (2025)
- [STAEformer: Spatio-Temporal Adaptive Embedding Makes Vanilla Transformer SOTA for Traffic Forecasting](https://mlsanity.com/paper/2308.10425/) (2023)
- [Multitask Learning for Network Traffic Classification](https://mlsanity.com/paper/1906.05248/) (2019)
- [TrafficFlowGAN: Physics-informed Flow based Generative Adversarial Network for Uncertainty Quantification](https://mlsanity.com/paper/2206.09319/) (2022)
- [Enhancing the Robustness via Adversarial Learning and Joint Spatial-Temporal Embeddings in Traffic Forecasting](https://mlsanity.com/paper/2208.03063/) (2022)
- [Dynamic Frequency Domain Graph Convolutional Network for Traffic Forecasting](https://mlsanity.com/paper/2312.11933/) (2023)
- [Human Mobility Prediction with Causal and Spatial-constrained Multi-task Network](https://mlsanity.com/paper/2206.05731/) (2022)
- [From Simulators to Digital Twins for Enabling Emerging Cellular Networks: A Tutorial and Survey](https://mlsanity.com/paper/2411.08907/) (2024)
- [Efficient Multi-Task Learning via Generalist Recommender](https://mlsanity.com/paper/2504.05318/) (2025)

## Built on by (most cited first, 1 of 1)

- [Cellular Predictions on the Move: What about Data?](https://mlsanity.com/paper/2606.25709/) (2026)

## Abstract (arXiv)

Network traffic forecasting plays a crucial role in intelligent network operations, but existing techniques often perform poorly when faced with limited data. Additionally, multi-task learning methods struggle with task imbalance and negative transfer, especially when modeling various service types. To overcome these challenges, we propose Sim-MSTNet, a multi-task spatiotemporal network traffic forecasting model based on the sim2real approach. Our method leverages a simulator to generate synthetic data, effectively addressing the issue of poor generalization caused by data scarcity. By employing a domain randomization technique, we reduce the distributional gap between synthetic and real data through bi-level optimization of both sample weighting and model training. Moreover, Sim-MSTNet incorporates attention-based mechanisms to selectively share knowledge between tasks and applies dynamic loss weighting to balance task objectives. Extensive experiments on two open-source datasets show that Sim-MSTNet consistently outperforms state-of-the-art baselines, achieving enhanced accuracy and generalization.