IJCAI 2026 | 时间序列(Time Series)论文总结【预测，分类，异常检测，表示学习，非平稳，不规则，流式时序以及LLM，通道策略等】

时空探索之旅

发布于 2026-06-12 16:19:36

7200

文章被收录于专栏：时空探索之旅时空探索之旅

IJCAI 2026在2026年8月15日至21日在德国不来梅（Bremen, Germany）举行。

本文总结了IJCAI 2026有关时间序列（Time Series）的相关文章，共计27篇，如有疏漏，欢迎补充。

时间序列Topic：预测，分类，异常检测，表示学习，非平稳时序，不规则时序，流式时序以及LLM，通道策略等

IJCAI2026 accepted papers: https://2026.ijcai.org/accepted-papers/

Main Track1. Scale-Invariant Conditional VAE for Coarse-Grained Economic Time-Series Forecasting2. EVENTTSF: Event-Aware Non-Stationary Time Series Forecasting3. Beyond Uniform Updates: Drift Pattern Aware Online Time Series Forecasting Under Delayed Feedback4. SeesawNet: Towards Non-stationary Time Series Forecasting with Balanced Modeling of Common and Specific Dependencies5. ConDyGNet: Constraint-Guided Dynamic Graph Networks for Multivariate Time Series Forecasting6. Perturbation Matters in Time Series Forecasting: A Wave-attention-aware Transformer Method7. Disentangling Coarse and Fine Latent Dynamics for Probabilistic Time Series Forecasting8. StreamMTS: Towards Streaming Multivariate Time Series Forecasting9. Attention as Selection: Semantic-Guided Time Series Forecasting10. SMLDR: Spectral Memory Learner with Dual-Retrieval for Time Series Forecasting11. StreamTimer: Efficient Inference for Long-Context Time Series Transformers12. From Values to Tokens: An LLM-Driven Framework for Context-Aware Time Series Forecasting via Symbolic Discretization13. Speaking Numbers to LLMs: Multi-Wavelet Number Embeddings for Time Series Forecasting14. FreSH: Frequency-Segmented Hierarchical Multi-Expert Framework for Multivariate Time Series Classification15. INSHAPE: Instance-Level Shapelets for Interpretable Time-Series Classification16. CASE-Net: Deep Spatio-Temporal Representation Learning via Causal Attention and Channel Recalibration for Multivariate Time Series Classification17. Multi-View Ensemble for Time Series Anomaly Detection via Coupling Flows18. AnoMamba: Aligning Reconstruction with Time Series Anomaly Detection via Selective Global Dependency Modeling19. Learning Hyperspherical Time–Frequency Representations for Time-Series Out-of-Distribution Detection20. H2SCAN: Adaptive Time Series Representation Learning via Heterogeneous Hypergraph Structure-aware Contrasts21. Frequency-Aware Augmentation and Alignment for Time Series Contrastive Learning22. Modeling Dynamic Mixtures of Time-Delay Systems from Streaming Time Series23. Efficient Time Series Clustering from Multiscale Reservoir Dynamics with Granular-Ball Anchoring Graph Optimization24. From Diversity to Uniformity: Cross-modal Time Series Modeling with Dependent Channel GroupingSpecial Track on AI4Tech: AI Enabling Critical Technologies25. Bridging the Data Scarcity in Venous Thromboembolism Detection: A Deep Learning Framework for Large-scale Irregular Clinical Time SeriesSurvey Track26. A Comprehensive Survey of Deep Learning for Multivariate Time Series Forecasting: A Channel Strategy Perspective27. From Time Series Analysis to Question Answering: A Survey in the LLM Era

Main Track

1 Scale-Invariant Conditional VAE for Coarse-Grained Economic Time-Series Forecasting

作者：Jianping Zhu, Lei Wang, Yang Chen, Bo Jin, Xiaopeng Wei

关键词：金融时间序列预测

2 EVENTTSF: Event-Aware Non-Stationary Time Series Forecasting

arXiv：https://arxiv.org/abs/2508.13434

代码：https://github.com/WinfredGe/EventTSF

作者：Yunfeng Ge, Ming Jin, Yiji Zhao, Hongyan Li, Bo Du, Chang Xu, Shirui Pan

关键词：非平稳时序预测，事件感知

3 Beyond Uniform Updates: Drift Pattern Aware Online Time Series Forecasting Under Delayed Feedback

作者：Xingwang Li, Fei Teng, Cong Zhou, Qiang Duan

关键词：在线时间序列预测

4 SeesawNet: Towards Non-stationary Time Series Forecasting with Balanced Modeling of Common and Specific Dependencies

arXiv：https://arxiv.org/abs/2605.14551

代码：https://github.com/dreamone-Lee/SeesawNet

作者：Hao Li, Lu Zhang, Liu Chong, Yankai Chen, Pengyang Wang, Yingjie Zhou

关键词：非平稳时序预测，实例归一化

5 ConDyGNet: Constraint-Guided Dynamic Graph Networks for Multivariate Time Series Forecasting

作者：Zhenzhou Li, Xiang Li, Zhibin Niu

关键词：多元时序预测，动态图

6 Perturbation Matters in Time Series Forecasting: A Wave-attention-aware Transformer Method

作者：Yiming Wang, Yiqing Su, Ximing Li, Changchun Li, Bing Wang

关键词：预测，Token注意力

7 Disentangling Coarse and Fine Latent Dynamics for Probabilistic Time Series Forecasting

代码：https://github.com/polars8948/COFE

作者：Changze Zhou, Ruichu Cai, Shengbin Nie, Juntao Fang, Jie Qiao, Zijian Li

关键词：概率时序预测

8 StreamMTS: Towards Streaming Multivariate Time Series Forecasting

作者：Binwu Wang, Jiaming Ma, Yudong Zhang, Pengkun Wang, Zhengyang Zhou, Xu Wang, Yang Wang

关键词：流式多元时序预测

9 Attention as Selection: Semantic-Guided Time Series Forecasting

代码：https://github.com/VIMLab-hfut/Attention-as-Selection

作者：Xueyu Luo, Qiang Lu, Sangui Jian, Yangxue Hu, Zhengyu Ying, Ye Yu, Wenxing Lu, Yan Qiao

关键词：语义驱动，预测

10 SMLDR: Spectral Memory Learner with Dual-Retrieval for Time Series Forecasting

作者：Zhenxin Li, Longquan Liao, Wenchang Zhang, Jiaying Zhang, Kaiwen Wei, Jiang Zhong, Linjiang Zheng

关键词：预测，双重检索，谱域

11 StreamTimer: Efficient Inference for Long-Context Time Series Transformers

作者：Xiyu Meng, Yuhan Wu, Canran Xiao, Yabo Dong, Duanqing Xu

关键词：预测，流式，自回归

12 From Values to Tokens: An LLM-Driven Framework for Context-Aware Time Series Forecasting via Symbolic Discretization

arXiv：https://arxiv.org/abs/2508.09191

代码：https://github.com/Xiaoyu-Tao/TokenCast

作者：Xiaoyu Tao, Shilong Zhang, Mingyue Cheng, Daoyu Wang, Tingyue Pan, Bokai Pan, Changqing Zhang, Shijin Wang

关键词：预测，LLM

13 Speaking Numbers to LLMs: Multi-Wavelet Number Embeddings for Time Series Forecasting

代码：https://github.com/DC-research/TempoWAVE

作者：Defu Cao, Zijie Lei, Muyan Weng, Jiao Sun, Yan Liu

关键词：预测，LLM，小波

14 FreSH: Frequency-Segmented Hierarchical Multi-Expert Framework for Multivariate Time Series Classification

代码：https://github.com/Wangmy2120/FreSH00

作者：Pingping Liu, Muyao Wang, Zijian Zhang, Tongshun Zhang, Hao Miao, Guorui Xie, Qingliang Li, Qiuzhan Zhou

关键词：分类，频域增强

15 INSHAPE: Instance-Level Shapelets for Interpretable Time-Series Classification

arXiv：https://arxiv.org/abs/2605.20088v1

作者：Seongjun Lee, Seokhyun Lee, Changhee Lee

关键词：分类，可解释性

16 CASE-Net: Deep Spatio-Temporal Representation Learning via Causal Attention and Channel Recalibration for Multivariate Time Series Classification

arXiv：https://arxiv.org/abs/2605.22043

作者：Fan Zhang, Yating Cui, Hua Wang

关键词：分类，时空表示学习，因果注意，通道重校准

17 Multi-View Ensemble for Time Series Anomaly Detection via Coupling Flows

作者：Wanghui Qiu, Chenxi Liu, Shiyan Hu, Zhengyu Li, Chenjuan Guo, Bin Yang

关键词：异常检测，耦合流

18 AnoMamba: Aligning Reconstruction with Time Series Anomaly Detection via Selective Global Dependency Modeling

作者：Junqi Chen, Xu Tan, Jie Chen, Susanto Rahardja

关键词：异常检测，mamba

19 Learning Hyperspherical Time–Frequency Representations for Time-Series Out-of-Distribution Detection

arXiv：https://arxiv.org/abs/2605.31155v1

代码：https://github.com/tiiuae/hypertf-time-series-ood

作者：Willian T. Lunardi, Samridha Shrestha, Martin Andreoni

关键词：分布外检测

20 SCAN: Adaptive Time Series Representation Learning via Heterogeneous Hypergraph Structure-aware Contrasts

作者：Biao Chen, Zijie Tang, Junhua Fang, Feng Lu, Lang Zhang, Pengpeng Zhao

关键词：表示学习，异质超图

21 Frequency-Aware Augmentation and Alignment for Time Series Contrastive Learning

作者：Yusen Liu, Zhichen Lai, Hua Lu, Xu Cheng, Xiufeng Liu, Huan Huo

关键词：对比学习，频域感知增强

22 Modeling Dynamic Mixtures of Time-Delay Systems from Streaming Time Series

arXiv：https://arxiv.org/abs/2605.26191v1

作者：Ren Fujiwara, Yasuko Matsubara, Yasushi Sakurai

关键词：流式时序建模

23 Efficient Time Series Clustering from Multiscale Reservoir Dynamics with Granular-Ball Anchoring Graph Optimization

作者：Yifan Wang, Lifeng Shen, Shuyin Xia, Yi Wang

关键词：聚类，锚点图优化策略

24 From Diversity to Uniformity: Cross-modal Time Series Modeling with Dependent Channel Grouping

代码：https://github.com/missCmj/TimeIG

作者：Minjun Cao, Hao Miao, Wentao Zhang, Senzhang Wang

关键词：跨模态时序建模，通道分组

Special Track on AI4Tech: AI Enabling Critical Technologies

25 Bridging the Data Scarcity in Venous Thromboembolism Detection: A Deep Learning Framework for Large-scale Irregular Clinical Time Series

作者：Can Xu, Runze Yang, Xinni Xiang, Yongtao Wu, Yaqin Huang, Haike Lei, Jie Yang

关键词：不规则临床时间序列

Survey Track

26 A Comprehensive Survey of Deep Learning for Multivariate Time Series Forecasting: A Channel Strategy Perspective

arXiv：https://arxiv.org/abs/2502.10721

作者：Xiangfei Qiu, Hanyin Cheng, Xingjian Wu, Junkai Lu, Jilin Hu, Chenjuan Guo, Christian S. Jensen, Bin Yang

关键词：多元时序预测，通道策略

27 From Time Series Analysis to Question Answering: A Survey in the LLM Era

arXiv：https://arxiv.org/abs/2506.11512v2

作者：Wei Li, Zhe Xie, Yuxuan Liang, Xinli Hao, Yunyao Cheng, Dan Pei, Xiaofeng Meng

关键词：时序分析，问答，LLM

此公众号的文章皆系本人原创，辛苦码字不易！如需转载，引用请注明出处。如商用联系作者。

如果觉得有帮助还请分享，在看，点赞

本文参与腾讯云自媒体同步曝光计划，分享自微信公众号。

原始发表：2026-06-10，如有侵权请联系 cloudcommunity@tencent.com 删除

series

本文分享自时空探索之旅微信公众号，前往查看

如有侵权，请联系 cloudcommunity@tencent.com 删除。

本文参与腾讯云自媒体同步曝光计划，欢迎热爱写作的你一起参与！

登录后参与评论

0 条评论

热度