References

A curated list of recent research and implementations using Mamba-based state space models.

Mamba-based Models and Applications

  1. Is Mamba Effective for Time Series Forecasting?
  2. STG-Mamba: Spatial-Temporal Graph Learning via Selective State Space Model
  3. Graph Mamba: Towards Learning on Graphs with State Space Models
    • Paper (arXiv:2402.08678)
    • Explores adapting state space models to graph data, with competitive results on graph classification tasks.
  4. Graph-Mamba: Towards Long-Range Graph Sequence Modeling with Selective State Spaces
  5. PointMamba: A Simple State Space Model for Point Cloud Analysis
  6. ClinicalMamba: A Generative Clinical Language Model on Longitudinal Clinical Notes

CILR 2025 Spatiotemporal Forecasting Papers (OpenReview)

  1. [Oral] High-Dynamic Radar Sequence Prediction for Weather Nowcasting Using Spatiotemporal Coherent Gaussian Representation
    • 🔗 Paper Link
    • Authors : Ziye Wang, Yiran Qin, Lin Zeng, Ruimao Zhang
    • Keywords : 3D Gaussian, dynamic reconstruction, radar forecasting, weather nowcasting
    • Score : 888
  2. [Spotlight] Learning Spatiotemporal Dynamical Systems from Point Process Observations
    • 🔗 Paper Link
    • Authors : Valerii Iakovlev, Harri Lähdesmäki
    • Keywords : Dynamics, spatiotemporal, neural, PDE, ODE
    • Score : 8688
  3. PhyMPGN: Physics-encoded Message Passing Graph Network for Spatiotemporal PDE Systems
    • 🔗 Paper Link
    • Authors : Bocheng Zeng et al.
    • Keywords : Physics encoding, spatiotemporal PDEs, graph networks, deep learning
    • TL;DR : Proposes a physics-encoded MPNN for modeling spatiotemporal PDE systems.
    • Score : 861088
  4. Expand and Compress: Exploring Tuning Principles for Continual Spatio-Temporal Graph Forecasting
    • 🔗 Paper Link
    • Authors : Wei Chen, Yuxuan Liang
    • Keywords : Spatiotemporal graph, continual forecasting, tuning
    • TL;DR : EAC uses expand and compress to tune prompt parameter pools efficiently.
    • Score : 3888
  5. WardropNet: Traffic Flow Predictions via Equilibrium-Augmented Learning
    • 🔗 Paper Link
    • Authors : Kai Jungel et al.
    • Keywords : Structured learning, combinatorial optimization, traffic equilibrium
    • Score : 8655
  6. Air Quality Prediction with Physics-Informed Dual Neural ODEs in Open Systems
    • 🔗 Paper Link
    • Authors : Jindong Tian et al.
    • Keywords : Air quality, physics-informed deep learning
    • TL;DR : A new physics-informed neural ODE method
    • Score : 66866
  7. Spectral-Refiner: Accurate Fine-Tuning of Spatiotemporal Fourier Neural Operator for Turbulent Flows
    • 🔗 Paper Link
    • Authors : Shuhao Cao et al.
    • Keywords : Operator learning, FNO, Navier-Stokes, CFD
    • TL;DR : Refines FNO using spectral methods for 10,000x accuracy.
    • Score : 5658
  8. Deep Random Features for Scalable Interpolation of Spatiotemporal Data
    • 🔗 Paper Link
    • Authors : Weibin Chen et al.
    • Keywords : Random features, deep Gaussian process, Bayesian deep learning, remote sensing
    • TL;DR : Scalable Bayesian DL for interpolating remote sensing data
    • Score : 388
  9. A Large-scale Dataset and Benchmark for Commuting Origin-Destination Flow Generation
    • 🔗 Paper Link
    • Authors : Can Rong et al.
    • Keywords : Commuting, OD flow, urban computing, weighted graph modeling
    • TL;DR : Provides OD matrix data from 3,000+ U.S. regions
    • Score : 5868
  10. Diffusion Transformer Captures Spatial-Temporal Dependencies: A Theory for Gaussian Process Data
    • 🔗 Paper Link
    • Authors : Hengyu Fu et al.
    • Keywords : Diffusion model, sequence data, spatial-temporal dependency
    • TL;DR : Theoretical analysis of Diffusion Transformer’s ability to capture spatiotemporal dependencies
    • Score : 5868
  11. PostCast: Generalizable Postprocessing for Precipitation Nowcasting via Unsupervised Blurriness Modeling
    • 🔗 Paper Link
    • Authors : Junchao Gong et al.
    • Keywords : SciAI, precipitation nowcasting, diffusion model
    • Score : 368
  12. VAE-Var: Variational Autoencoder-Enhanced Variational Methods for Data Assimilation in Meteorology
    • 🔗 Paper Link
    • Authors : Yi Xiao et al.
    • Keywords : Data assimilation, VAE, weather forecasting
    • Score : 886665
  13. WeatherGFM: Learning a Weather Generalist Foundation Model via In-context Learning
    • 🔗 Paper Link
    • Authors : Xiangyu Zhao et al.
    • Keywords : SciAI, weather foundation model, in-context learning
    • Score : 66310
  14. HR-Extreme: A High-Resolution Dataset for Extreme Weather Forecasting
    • 🔗 Paper Link
    • Authors : Nian Ran et al.
    • Keywords : Weather dataset, extreme weather, NWP
    • TL;DR : HR-Extreme enables evaluation of extreme weather forecasting
    • Score : 5868
  15. Continuous Ensemble Weather Forecasting with Diffusion Models
    • 🔗 Paper Link
    • Authors : Martin Andrae et al.
    • Keywords : Weather forecasting, diffusion, ensemble prediction
    • TL;DR : New diffusion-based ensemble forecasting method
    • Score : 5555
  16. CirT: Global Subseasonal-to-Seasonal Forecasting with Geometry-inspired Transformer
    • 🔗 Paper Link
    • Authors : Yang Liu et al.
    • Keywords : Weather and climate forecasting
    • TL;DR : Introduces a geometry-inspired Transformer for subseasonal-to-seasonal forecasting
    • Score : 666

ICDM 2024 Spatiotemporal Learning and Applications Papers

  1. Towards Efficient Ridesharing via Order-Vehicle Pre-Matching Using Attention Mechanism
    • 🔗 PDF Link
    • Authors : Zhidan Liu, Jinye Lin, Zhiyu Xia, Chao Chen, and Kaishun Wu
    • Keywords : Ridesharing, Order-vehicle pre-matching, Self-attention mechanism, Spatiotemporal matching
  1. LISA: Learning-Integrated Space Partitioning Framework for Traffic Accident Forecasting on Heterogeneous Spatiotemporal Data
    • Authors : Bang An, Xun Zhou, Amin Khezerlou, Nick Street, Jinping Guan, and Jun Luo
    • Keywords : Traffic accident forecasting, Spatiotemporal data mining
  1. Align Along Time and Space: A Graph Latent Diffusion Model for Traffic Dynamics Prediction
    • Authors : Yuhang Liu, Yingxue Zhang, Xin Zhang, Yu Yang, Yiqun Xie, Sahar Ghanipoor Machiani, Yanhua Li, and Jun Luo
    • Keywords : Diffusion models, Urban dynamics prediction, Latent diffusion, Spatiotemporal modeling
  1. Traffic Pattern Sharing for Federated Traffic Flow Prediction with Personalization
    • Authors : Hang Zhou, Wentao Yu, Sheng Wan, Yongxin Tong, Tianlong Gu, and Chen Gong
    • Keywords : Spatiotemporal data, Traffic flow prediction, Personalized federated learning
  1. MetaSTC: A Meta Spatio-Temporal Learning Paradigm for Traffic Flow Prediction (Short Paper)
    • Authors : Kexin Xu, Zhemeng Yu, Yucen Gao, Songjian Zhang, Jun Fang, Xiaofeng Gao, and Guihai Chen
    • Keywords : Spatiotemporal data mining, Meta-learning, Traffic flow prediction, Backbone-agnostic design
  1. 2DXformer: Dual Transformers for Wind Power Forecasting with Dual Exogenous Variables
    • Authors : Yajuan Zhang, Jiahai Jiang, Yule Yan, Liang Yang, and Ping Zhang
    • Keywords : Wind power forecasting, Spatiotemporal forecasting, Exogenous variables, Variable correlation
  1. Futures Quantitative Investment with Heterogeneous Continual Graph Neural Network
    • Authors : Zhizhong Tan, Min Hu, Bin Liu, and Guosheng Yin
    • Keywords : Continual learning, Futures price forecasting, Graph neural networks, Spatiotemporal data
  1. Controllable Visit Trajectory Generation with Spatiotemporal Constraints
    • Authors : Yuting Qiang, Jianbin Zheng, Lixia Wu, Haomin Wen, Junhong Lou, and Minhui Deng
    • Keywords : Cross-modal learning, Contrastive learning, Query-POI matching, Spatiotemporal constraints
  1. A Momentum Contrastive Learning Framework for Query-POI Matching
    • Authors : Haowen Lin, John Krumm, Cyrus Shahabi, and Li Xiong
    • Keywords : Trajectory generation, Spatiotemporal systems, Controlled generation, POI matching
  1. (Demo) VIA AI: Reliable Deep Reinforcement Learning for Traffic Signal Control
    • Authors : Matvey Gerasyov, Dmitrii Kiselev, Maxim Beketov, and Ilya Makarov
    • Keywords : Traffic signal control, Deep reinforcement learning, Urban traffic optimization