2026
Learning to Factorize and Adapt: A Versatile Approach Toward Universal Spatio-Temporal Foundation Models
Siru Zhong, Junjie Qiu, Yangyu Wu, Yiqiu Liu, Yuanpeng He, Zhongwen Rao, Bin Yang, Chenjuan Guo, Hao Xu, Yuxuan Liang# (# corresponding author)
Submitted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). Under review.
We extend FactoST to transfer more easily across domains and make predictions over flexible time horizons, even when little or no target-domain data is available.
Learning to Factorize and Adapt: A Versatile Approach Toward Universal Spatio-Temporal Foundation Models
Siru Zhong, Junjie Qiu, Yangyu Wu, Yiqiu Liu, Yuanpeng He, Zhongwen Rao, Bin Yang, Chenjuan Guo, Hao Xu, Yuxuan Liang# (# corresponding author)
Submitted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). Under review.
We extend FactoST to transfer more easily across domains and make predictions over flexible time horizons, even when little or no target-domain data is available.
2025
Learning to Factorize Spatio-Temporal Foundation Models
Siru Zhong, Junjie Qiu, Yangyu Wu, Xingchen Zou, Bin Yang, Chenjuan Guo, Hao Xu, Yuxuan Liang# (# corresponding author)
Neural Information Processing Systems (NeurIPS) 2025 Spotlight
Spatio-temporal forecasting predicts how conditions change across both space and time—for example, traffic at different locations in a city. We build a model that learns recurring patterns from many datasets, then adapts efficiently to a new city or application.
Learning to Factorize Spatio-Temporal Foundation Models
Siru Zhong, Junjie Qiu, Yangyu Wu, Xingchen Zou, Bin Yang, Chenjuan Guo, Hao Xu, Yuxuan Liang# (# corresponding author)
Neural Information Processing Systems (NeurIPS) 2025 Spotlight
Spatio-temporal forecasting predicts how conditions change across both space and time—for example, traffic at different locations in a city. We build a model that learns recurring patterns from many datasets, then adapts efficiently to a new city or application.