ECML PKDD 2025 | 时空数据（Spatial-Temporal）论文总结

ECML PKDD是CCF B类会议。ECML PKDD2025将在2025年9月15号到19号在加葡萄牙波尔图（ Porto, Portugal）举行，本文总结了ECML PKDD2025有关时空数据（Spatial-Temporal）相关文章，共计10篇，其中1-6为Research Track，7-10为ADS Track。

时空数据Topic：地理基础模型，时空预测，城市区域表示学习，空间插值，交通事故分类，天气预报等。如有疏漏，欢迎补充！

Research

1 Fine-tune Smarter, Not Harder: Parameter-Efficient Fine-Tuning for Geospatial Foundation Models

代码：https://github.com/IBM/peft-geofm

作者：Francesc Marti Escofet (IBM Research Europe); Benedikt Blumenstiel (IBM Research Europe)*; Linus Scheibenreif (University of St. Gallen); Paolo Fraccaro (IBM Research Europe); Konrad Schindler (ETH Zurich)

关键词：地理基础模型，地球观测，参数高效微调

2 Hierarchical Information-Guided Spatio-Temporal Mamba for Stock Time Series Forecasting

作者：Wenbo Yan (Peking University)*; Shurui Wang (Peking University); Ying Tan (Peking University)

关键词：股票时序预测，时空Mamba

3 ST-LoRA: Low-rank Adaptation for Spatio-temporal Forecasting

代码：https://github.com/RWLinno/ST-LoRA

作者：Weilin Ruan (The Hong Kong University of Science and Technology (Guangzhou)); Wei Chen (The Hong Kong University of Science and Technology (Guangzhou)); Xilin Dang (The Chinese University of Hong Kong); Jianxiang Zhou (The Hong Kong University of Science and Technology (Guangzhou)); Weichuang Li (The Hong Kong University of Science and Technology (Guangzhou)); Xu Liu (National University of Singapore); Yuxuan Liang (The Hong Kong University of Science and Technology (Guangzhou))*

关键词：时空预测，LoRA

4 C3DE: Causal-Aware Collaborative Neural Controlled Differential Equation for Long-Term Urban Crowd Flow Prediction

代码：https://github.com/Sonder-arch/C3DE

作者：Yuting Liu (Nanjing University of Aeronautics and Astronautics); Qiang Zhou ( Nanjing University of Aeronautics and Astronautics)*; Hanzhe Li ( Nanjing University of Aeronautics and Astronautics); Chenqi Gong (Chongqing University); Jingjing Gu (Nanjing University of Aeronautics and Astronautics)

关键词：城市流量预测，神经可控微分方程，反事实推理

5 GraphJCL: A Dual-Perspective Graph-Based Framework for Urban Region Representation via Joint Contrastive Learning

作者：Yaya Zhao (Center for Applied Statistics, School of Statistics, Renmin University of China)*; Kaiqi Zhao (The University of Auckland); Zixuan Tang (Center for Applied Statistics, Renmin University of China); Xiaoling Lu (Center for Applied Statistics, Renmin University of China); Yuanyuan Zhang (Beijing Baixingkefu Network Technology Co., Ltd.); Yalei Du ( Beijing Baixingkefu Network Technology Co., Ltd.)

关键词：城市区域表示，图神经网络，对比学习

6 Memory-enhanced Invariant Prompt Learning for Urban Flow Prediction under Distribution Shifts

作者：Haiyang Jiang (The University of Queensland); Tong Chen (The University of Queensland)*; Wentao Zhang (Peking University); Quoc Viet Hung Nguyen (Griffith University); Yuan Yuan (Tsinghua University); Yong Li (Tsinghua University); Hongzhi Yin (h.yin1@uq.edu.au)

关键词：时空图神经网络，分布外泛化，不变学习

ADS Track

7 CESI: Sparse Input Spatial Interpolation for Heterogeneous and Noisy Hybrid Wireless Sensor Networks

作者：Chaofan Li (Karlsruhe Institute of Technology)*; Till Riedel (Karlsruhe Institute of Technology); Michael Beigl (Karlsruhe Institute of Technology)

关键词：空间插值

8 Enhancing Traffic Accident Classifications: Application of NLP Methods for City Safety

代码：https://github.com/enesozeren/enhancing-traffic-accident-classifications

作者：Enes Oezeren (LMU Munich); Alexander Ulbrich (LMU Munich); Sascha Filimon (City of Munich); David Ruegamer (LMU Munich); Andreas Bender (LMU Munich)*

关键词：事故分类，少样本学习，主题模型

9 Progressive Decomposition-enhanced Time-Varying Graph Neural Network for Traffic Forecasting

作者：Jianuo Ji (Harbin Engineering University)*; Hongbin Dong (Harbin Engineering University); Xiaoping Zhang (China Academy of Chinese Medical Sciences)

关键词：交通预测，图神经网络

10 Low Visibility Forecasting Using Numerical Weather Prediction Data

作者：Topon Paul (Toshiba Corporation)*; Vidhisha Reddy (Toshiba Software (India) Pvt. Ltd); Sai Prem Kumar Ayyagari (Toshiba Software (India) Pvt. Ltd); Ryusei Shingaki (Toshiba Corporation); Kaneharu Nishino (Toshiba Corporation); Yoshiaki Shiga (Toshiba Corporation)

关键词：低能见度预测，天气预报数据

相关链接

ECML PKDD 2025 preprint：https://ecmlpkdd.org/preprints/2025/

欢迎各位作者投稿近期有关时空数据和时间序列录用的顶级会议和期刊的优秀文章解读，我们将竭诚为您宣传,共同学习进步。如有意愿,请通过后台私信与我们联系。