Smart City Gnosys

Smart city article details

Title	Coms2T: A Complementary Spatiotemporal Learning System For Data-Adaptive Model Evolution
ID_Doc	15460
Authors	Zhou Z.; Huang Q.; Wang B.; Hou J.; Yang K.; Liang Y.; Zheng Y.; Wang Y.
Year	2025
Published	IEEE Transactions on Pattern Analysis and Machine Intelligence
DOI	http://dx.doi.org/10.1109/TPAMI.2025.3576805
Abstract	Spatiotemporal (ST) learning has become a crucial technique to enable smart cities and sustainable urban development. Current ST learning models capture the heterogeneity via various spatial convolution and temporal evolution blocks. However, rapid urbanization leads to fluctuating distributions in urban data and city structures, resulting in existing methods suffering generalization and data adaptation issues. Despite efforts, existing methods fail to deal with newly arrived observations, and the limitation of those methods with generalization capacity lies in the repeated training that leads to inconvenience, inefficiency and resource waste. Motivated by complementary learning in neuroscience, we introduce a prompt-based complementary spatiotemporal learning termed ComS2T, to empower the evolution of models for data adaptation. We first disentangle the neural architecture into two disjoint structures, a stable neocortex for consolidating historical memory, and a dynamic hippocampus for new knowledge update. Then we train the dynamic spatial and temporal prompts by characterizing distribution of main observations to enable prompts adaptive to new data. This data-adaptive prompt mechanism, combined with a two-stage training process, facilitates fine-tuning of the neural architecture conditioned on prompts, thereby enabling efficient adaptation during testing. Extensive experiments validate the efficacy of ComS2T in adapting various spatiotemporal out-of-distribution scenarios while maintaining effective inferences. © 1979-2012 IEEE.
Author Keywords	Complementary learning system; OOD generalization; spatiotemporal learning; urban computing

Similar Articles

Id	Similarity	Authors	Title	Published
60280	0.884	Li Z.; Xia L.; Tang J.; Xu Y.; Shi L.; Xia L.; Yin D.; Huang C.	Urbangpt: Spatio-Temporal Large Language Models	Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2024)
61766	0.873	Fang Z.; Wu D.; Pan L.; Chen L.; Gao Y.	When Transfer Learning Meets Cross-City Urban Flow Prediction: Spatio-Temporal Adaptation Matters	IJCAI International Joint Conference on Artificial Intelligence (2022)
44815	0.869	Chen J.; Liu T.; Li R.	Region Profile Enhanced Urban Spatio-Temporal Prediction Via Adaptive Meta-Learning	International Conference on Information and Knowledge Management, Proceedings (2023)
43449	0.868	Zhang Z.; Zhao X.; Liu Q.; Zhang C.; Ma Q.; Wang W.; Zhao H.; Wang Y.; Liu Z.	Promptst: Prompt-Enhanced Spatio-Temporal Multi-Attribute Prediction	International Conference on Information and Knowledge Management, Proceedings (2023)
34890	0.865	Yao H.; Liu Y.; Wei Y.; Tang X.; Li Z.	Learning From Multiple Cities: A Meta-Learning Approach For Spatial-Temporal Prediction	The Web Conference 2019 - Proceedings of the World Wide Web Conference, WWW 2019 (2019)
13501	0.864	Li S.; Li H.; Li X.; Xu Y.; Lin Z.; Jiang H.	Causal Intervention Is What Large Language Models Need For Spatio-Temporal Forecasting	IEEE Transactions on Cybernetics (2025)
11537	0.857	Li T.; Zhang J.; Bao K.; Liang Y.; Li Y.; Zheng Y.	Autost: Efficient Neural Architecture Search For Spatio-Temporal Prediction	Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2020)
17754	0.857	Hu J.; Liang Y.; Fan Z.; Liu L.; Yin Y.; Zimmermann R.	Decoupling Long- And Short-Term Patterns In Spatiotemporal Inference	IEEE Transactions on Neural Networks and Learning Systems, 35, 11 (2024)
11538	0.856	Zhang Z.; Zhao X.; Miao H.; Zhang C.; Zhao H.; Zhang J.	Autostl: Automated Spatio-Temporal Multi-Task Learning	Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023, 37 (2023)
52571	0.856	Wang S.; Miao H.; Li J.; Cao J.	Spatio-Temporal Knowledge Transfer For Urban Crowd Flow Prediction Via Deep Attentive Adaptation Networks	IEEE Transactions on Intelligent Transportation Systems, 23, 5 (2022)