KDD 2026于2026年8月9日至13日于韩国济州（Jeju, Korea ）正在举行。
本文总结了KDD 2026（February Cycle）上有关时间序列（Time Series）的相关论文。
注：由于时间序列的论文高达57篇，笔者将分为上下2篇推文来总结，下篇主要涉及表示学习，生成，异常检测和分类等论文，包含Research，Applied Data Science，AI for Science和Datasets and Benchmarks 4个Track的论文，共计29篇。
时间序列Topic：表示学习，生成，异常检测，分类，因果推断，RAG，Benchmark，（工业，金融，医疗）时序等。
Research
1 Estimating Mutual Information between Time Series and Temporal Event Sequences Across Diverse Analysis Tasks
链接：https://dl.acm.org/doi/10.1145/3770855.3817693
代码：https://github.com/HaojiHu/Multimodal-Temporal-Data-Quantification
作者：Haoji Hu (University of Minnesota – Twin Cities); Huaqing Mao (University of Minnesota – Twin Cities); Yijun Lin (University of Minnesota – Twin Cities); Xiaowei Jia (University of Pittsburgh); Jinwei Zhou (University of Minnesota – Twin Cities); Minoh Jeong (Inha University); Yao-Yi Chiang (University of Minnesota – Twin Cities)
关键词：互信息；多模态时序数据；时间序列；时序事件序列；季节性；特征选择；协变量
2 Parallel Complex Diffusion for Scalable Time Series Generation
链接：https://dl.acm.org/doi/10.1145/3770855.3817791
代码：https://github.com/RongyaoCai/PaCoDi
作者：Rongyao Cai (Institute of Cyber-Systems and Control, Zhejiang University); Yuxi Wan (Institute of Cyber-Systems and Control, Zhejiang University); Kexin Zhang (Institute of Cyber-Systems and Control, Zhejiang University); Ming Jin (School of Information and Communication Technology, Griffith University); Zhiqiang Ge (School of Mathematics, Southeast University); Qingsong Wen (Squirrel Ai Learning); Yong Liu (Institute of Cyber-Systems and Control, Zhejiang University)
关键词：时间序列；频谱扩散理论；计算复杂度
3 Latent Causal Flow: Continuous Backdoor Adjustment for Time Series Generation
链接：https://dl.acm.org/doi/10.1145/3770855.3818059
代码：https://github.com/decisionintelligence/LCF
作者：Changchen Song (East China Normal University); Peng Chen (East China Normal University); Yang Shu (East China Normal University)
关键词：时间序列生成；因果推断；流匹配；潜变量模型；后门调整
4 PolarFormer: Radial-Angular Latent Modeling for Unconditional Time Series Generation
链接：https://dl.acm.org/doi/10.1145/3770855.3817753
代码：https://github.com/decisionintelligence/PolarFormer
作者：Jiahong Lyu (School of Data Science and Engineering, East China Normal University); Hongfan Gao (School of Data Science and Engineering, East China Normal University); Wangmeng Shen (School of Data Science and Engineering, East China Normal University); Qingsong Zhong (School of Data Science and Engineering, East China Normal University); Jilin Hu (School of Data Science and Engineering, East China Normal University)
关键词：时间序列生成；向量量化变分自编码器；Transformer；极分解；混合专家模型
5 DAMR: Dual Adaptive Multi-Head Representation Learning for Multivariate Time Series Anomaly Detection
链接：https://dl.acm.org/doi/10.1145/3770855.3817943
代码：https://github.com/wyn404/DAMR
作者：Yining Wang (Shandong University); Fujun Han (The Chinese University of Hong Kong, Shenzhen); Ke Li (Huazhong University of Science and Technology); Shuhan Liu (Shandong University); Tao Chen (Chongqing University of Science and Technology); Zhaohui Peng (Shandong University)
关键词：多变量时间序列；异常检测；对比学习；相似性
6 TimeRadar: A Domain-Rotatable Foundation Model for Time Series Anomaly Detection
链接：https://dl.acm.org/doi/10.1145/3770855.3818062
代码：https://github.com/mala-lab/TimeRadar
作者：Hui He (School of Computing and Information Systems, Singapore Management University); Hezhe Qiao (School of Computing and Information Systems, Singapore Management University); Yutong Chen (Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences); Kun Yi (State Information Center); Guansong Pang (School of Computing and Information Systems, Singapore Management University)
关键词：时间序列；基础模型；异常检测
7 Bridging Classification and Reconstruction: Cooperative Time Series Anomaly Detection
链接：https://dl.acm.org/doi/10.1145/3770855.3818108
代码：https://github.com/richard-tang199/CoAD
作者：Qideng Tang (National Key Laboratory of Information Systems Engineering, National University of Defense Technology,Zhejiang Key Laboratory of Space Information Sensing and Transmission, Hangzhou Dianzi University); Chaofan Dai (National Key Laboratory of Information Systems Engineering, National University of Defense Technology); Wubin Ma (National Key Laboratory of Information Systems Engineering, National University of Defense Technology); Yahui Wu (National Key Laboratory of Information Systems Engineering, National University of Defense Technology); Haohao Zhou (National Key Laboratory of Information Systems Engineering, National University of Defense Technology); Tao Zhang (College of Systems Engineering, National University of Defense Technology); Huan Li (Zhejiang University); Dalin Zhang (Zhejiang Key Laboratory of Space Information Sensing and Transmission, Hangzhou Dianzi University)
关键词：时间序列；异常检测；掩码自编码器；异常样本暴露
8 SEA-FGT: Frequency-Guided Transformer with Semantic Expert Augment for Time Series Anomaly Detection
链接：https://dl.acm.org/doi/10.1145/3770855.3817823
作者：Wei Huang (College of Computer and Data Science, Fuzhou University,Engineering Research Center of Big Data Intelligence, Ministry of Education, Fuzhou University); Zhihong Wang (College of Computer and Data Science, Fuzhou University); Yanyong Huang (Joint Laboratory of Data Science and Business Intelligence, Southwestern University of Finance and Economics); Jia Liu (School of Computer and Software Engineering, Xihua University); Xiaocao Ouyang (School of Computing and Artificial Intelligence, Southwestern University of Finance and Economics)
关键词：多变量时间序列异常检测；表征学习；语义增强；频域建模
9 LEFT: Learnable Fusion of Tri-view Tokens for Unsupervised Time Series Anomaly Detection
链接：https://dl.acm.org/doi/10.1145/3770855.3818044
代码：https://github.com/DezhengWang/Left.git
作者：Dezheng Wang (School of Automation, Southeast University); Tong Chen (School of Electrical Engineering and Computer Science, The University of Queensland); Guansong Pang (School of Computing and Information Systems, Singapore Management University); Congyan Chen (School of Automation, Southeast University); Shihua Li (School of Automation, Southeast University); Hongzhi Yin (School of Electrical Engineering and Computer Science, The University of Queensland)
关键词：时间序列异常检测；无监督学习；跨视图一致性；三视图标记化；可学习滤波器组
10 LatentFlow: Discovering Latent Continuous Dynamics across Channels for Multivariate Time Series Anomaly Detection
链接：https://dl.acm.org/doi/10.1145/3770855.3818034
作者：Lijun Sun (Beihang University,Zhongguancun Laboratory); Shuai Zhang (Zhongguancun Laboratory); Xin Xue (Beihang University); Lanhao Li (Beihang University); Haoyi Zhou (Beihang University,Zhongguancun Laboratory); Jianxin Li (Beihang University,Zhongguancun Laboratory)
关键词：时间序列异常检测；动态依赖建模
11 CAAD: Causality-Aware Multivariate Time Series Anomaly Detection via Multi-Scale Alignment and Structural Causal Consistency
链接：https://dl.acm.org/doi/10.1145/3770855.3817978
作者：Xin Wang (Stony Brook University); Yunshi Wen (Rensselaer Polytechnic Institute); Yanan He (Yale University); Haotian Xu (Stony Brook University); Youlan Zhao (Stony Brook University); Michel Ferreira Cardia Haddad (Queen Mary, University of London); Tengfei Ma (Stony Brook University)
关键词：时间序列；异常检测；因果性；多尺度对齐
12 Rethinking Time Series Anomaly Detection from a Dynamic Perspective: Temporal-Frequency-Curvature Fusion
链接：https://dl.acm.org/doi/10.1145/3770855.3817936
代码：https://github.com/CSTCloudOps/TFC.
作者：Hang Cui (Computer Network Information Center, Chinese Academy of Sciences,University of Chinese Academy of Sciences); Zexin Wang (Computer Network Information Center, Chinese Academy of Sciences); Changhua Pei (Computer Network Information Center, Chinese Academy of Sciences,Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences); Juncheng Hu (Jilin University); Haotian Si (Computer Network Information Center, Chinese Academy of Sciences); Quan Zhou (Computer Network Information Center, Chinese Academy of Sciences); Cenjie Hu (Shenyang Institute of Automation, Chinese Academy of Sciences); Jingjing Li (Computer Network Information Center, Chinese Academy of Sciences); Dan Pei (Tsinghua University); Gaogang Xie (Computer Network Information Center, Chinese Academy of Sciences)
关键词：时间序列异常检测；深度学习
13 ReCATS: Replay-Free Continual Anomaly Detection for Non-Stationary Multivariate Time Series
链接：https://dl.acm.org/doi/10.1145/3770855.3817985
代码：https://github.com/Li-Qiuyang/ReCATS
作者：Qiuyang Li (Dalian Martime University); Qian Ma (Dalian Martime University); Zhongming Yao (Aalborg University); Shikai Guo (Dalian Maritime University); Yu Gu (Northeastern University); Ge Yu (Northeastern University); Tianyi Li (Aalborg University)
关键词：多变量时间序列；异常检测；持续学习
14 Federated Domain Generalization for Time-Series Classification via Dynamics-to-Domain Generation
链接：https://dl.acm.org/doi/10.1145/3770855.3817676
作者：Haoran Shi (Nanyang Technological University); Junru Zhang (Zhejiang University); Xiaoli Tang (Nanyang Technological University); Yifei Zhang (Northwest Polytechnical University); Hao Zhu (Data61&CSIRO); Yabo Dong (Zhejiang University); Lizhen Cui (Shandong University); Han Yu (Nanyang Technological University)
关键词：时间序列分类；联邦学习；领域泛化；扩散模型
15 Extracting Explainable Temporal Features in Multivariate Time Series Classification Pipelines
链接：https://dl.acm.org/doi/10.1145/3770855.3818014
代码：https://doi.org/10.5281/zenodo.20507211
作者：Ido Ikar (Bar-Ilan University); Amit Somech (Bar-Ilan University)
关键词：时间序列分类；可解释性；特征提取
16 AnchorMoE: Interpretable Time Series Classification via Anchor-Routed MoE
链接：https://dl.acm.org/doi/10.1145/3770855.3817786
代码：https://github.com/kuxit/AnchorMoE
作者：Tao Xie (Guangdong University of Technology); Zexi Tan (Guangdong University of Technology); Haoyi Xiao (Guangdong University of Technology); Mengke Li (Shenzhen University); Yiqun Zhang (Guangdong University of Technology); Yang Lu (Xiamen University); Cuie Yang (Northeastern University); Yiu-ming Cheung (Hong Kong Baptist University)
关键词：多变量时间序列分类；可解释机器学习；自解释模型；混合专家模型
17 Continuous Causal Component and Structure Discovery from Time Series
链接：https://dl.acm.org/doi/10.1145/3770855.3818046
作者：Dezhi Yang (School of Software, Shandong University); Jun Wang (Joint SDU-NTU Centre for Artificial Intelligence Research, Shandong University); Carlotta Domeniconi (Department of Computer Science, George Mason University); Dong Wu (Inspur Data Technology Co, Ltd); Dong Zhang (Inspur Data Technology Co, Ltd); Jinglin Zhang (School of Control Science and Engineering, Shandong University); Guoxian Yu (School of Software, Shandong University)
关键词：时序因果发现；连续因果成分；双稀疏性诱导；神经常微分方程
18 TRIAD: Ternary Information Routing for Multimodal Time-Series Sensing Signals
链接：https://dl.acm.org/doi/10.1145/3770855.3818194
代码：https://doi.org/10.5281/zenodo.20405997
作者：Yusen Liu (University of Technology Sydney); Shouxing Ma (University of Technology Sydney); Hua Lu (Aalborg University); Xu Cheng (Tianjin University of Technology); Xiufeng Liu (Technical University of Denmark); Huan Huo (University of Technology Sydney)
关键词：多模态学习；时间序列；自监督学习；解耦表征；传感器融合
19 Structure-Aware Abstraction of Hierarchical Time Series
链接：https://dl.acm.org/doi/10.1145/3770855.3817713
作者：Yihan Wu (Shanghai Jiao Tong University); Xuliang Zhu (Antai College of Economics and Management, Shanghai Jiao Tong University); Guozhong Li (KAUST); Kai Wang (Antai College of Economics and Management, Shanghai Jiao Tong University); Xuemin Lin (School of Data Science, The Chinese University of Hong Kong, Shenzhen)
关键词：层次时间序列；时间序列抽象；时间序列总结；层次化摘要
20 TimeBlocks: Versatile and Continual Time-Series Blockbase
链接：https://dl.acm.org/doi/10.1145/3770855.3817919
代码：https://doi.org/10.6084/m9.figshare.32399505
作者：David Campos (Aalborg University,University of Amsterdam); Bin Yang (East China Normal University); Tung Kieu (Aalborg University); Lei Chen (Hong Kong University of Science and Technology); Chenjuan Guo (East China Normal University); Christian S. Jensen (Aalborg University)
关键词：时间序列；高效机器学习
21 Directional Time Series Editing via Retrieval-Guided Jacobian-Vector Inference
链接：https://dl.acm.org/doi/10.1145/3770855.3817865
代码：https://doi.org/10.5281/zenodo.20498284
作者：Yifan Bao (National University of Singapore); Yihao Ang (National University of Singapore); Qiang Huang (Harbin Institute of Technology (Shenzhen)); Anthony K. H. Tung (National University of Singapore); Zhiyong Huang (National University of Singapore)
关键词：时间序列编辑；时间序列生成；雅可比-向量推理；检索引导学习
22 Adaptive Prototypical Contrastive Learning for Time Series Clustering
链接：https://dl.acm.org/doi/10.1145/3770855.3817773
代码：https://github.com/William-Liwei/APCL
作者：Wei Li (School of Computer Engineering and Science, Shanghai University)
关键词：时间序列；自监督学习；对比学习；聚类；表征学习
Applied Data Science
23 Retrieval-Augmented Generation with Covariate Time Series
链接：https://dl.acm.org/doi/10.1145/3770855.3818471
作者：Kenny Ye Liang (Tsinghua University); Zhongyi Pei (Tsinghua University); Huan Zhang (China Southern Airlines Co., Ltd.); Yuhui Liu (China Southern Airlines Co., Ltd.); Shaoxu Song (BNRist, Tsinghua University); Jianmin Wang (Tsinghua University)
关键词：时间序列基础模型；检索增强生成；时间序列预测；Apache IoTDB
24 Aura: Universal Multi-dimensional Exogenous Integration for Aviation Time Series
链接：https://dl.acm.org/doi/10.1145/3770855.3818321
作者：Jiafeng Lin (School of Software, BNRist, Tsinghua University); Mengren Zheng (Hongshen College, Chongqing University); Simeng Ye (School of Software, Tsinghua University); Yuxuan Wang (School of Software, BNRist, Tsinghua University); Huan Zhang (China Southern Airlines Co., Ltd.); Yuhui Liu (China Southern Airlines Co., Ltd.); Zhongyi Pei (School of Software, BNRist, Tsinghua University); Jianmin Wang (School of Software, BNRist, Tsinghua University)
关键词：时间序列；外生变量；多模态
Datasets and Benchmarks
25 TEFD: A Benchmark for Natural Language to Flux Query Generation in Time-Series Databases
链接：https://dl.acm.org/doi/10.1145/3770855.3817508
代码：https://github.com/gta886/TEFD-Benchmark
作者：Xuefeng Wu (College of Computer Science and Software Engineering, Shenzhen University); Yuanfeng Song (Independent Researcher); Jiawei Wen (College of Computer Science and Software Engineering, Shenzhen University); Dandan Huang (College of Computer Science and Software Engineering, Shenzhen University); Raymond Chi-Wing Wong (Department of Computer Science and Engineering (CSE), The Hong Kong University of Science and Technology); Haodi Zhang (College of Computer Science and Software Engineering, Shenzhen University)
关键词：时间序列数据库；文本到 Flux 查询；自然语言接口；数据集生成；InfluxDB
26 TSRBench: Benchmarking Time-Series Retrieval
链接：https://dl.acm.org/doi/10.1145/3770855.3817598
代码：https://adeval.cstcloud.cn/tsr
作者：Cenjie Hu (Shenyang Institute of Automation, Chinese Academy of Sciences,University of the Chinese Academy of Sciences); Hang Cui (Computer Network Information Center, Chinese Academy of Sciences); Zexin Wang (Computer Network Information Center, Chinese Academy of Sciences); Juncheng Bao (Shenyang Institute of Automation, Chinese Academy of Sciences); Jingwen Yang (Computer Network Information Center, Chinese Academy of Sciences); Jingjing Li (Computer Network Information Center, Chinese Academy of Sciences); Changhua Pei (Computer Network Information Center, Chinese Academy of Sciences,Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences); Dan Pei (Tsinghua University); Gaogang Xie (Computer Network Information Center, Chinese Academy of Sciences)
关键词：时间序列检索；深度学习
27 FinTexTS: Financial Text-Paired Time-Series Dataset via Semantic-Based and Multi-Level Pairing
链接：https://dl.acm.org/doi/10.1145/3770855.3817468
代码：https://github.com/leejaehoon2016/FinTexTS
作者：Jaehoon Lee (LG AI Research); Suhwan Park (Ulsan National Institute of Science and Technology); Taeyoon Lim (LG AI Research); Seunghan Lee (LG AI Research); Jun Seo (LG AI Research); Dongwan Kang (LG AI Research); Hwanil Choi (LG AI Research); Minjae Kim (LG AI Research); Sungdong Yoo (LG AI Research); Soonyoung Lee (LG AI Research); Yongjae Lee (Ulsan National Institute of Science and Technology); Wonbin Ahn (LG AI Research)
关键词：股票预测；金融多模态数据集；时间序列
AI for Sciences
28 MedTVL: Harnessing Vision and Language for Medical Time Series Classification
链接：https://dl.acm.org/doi/10.1145/3770855.3818883
代码：https://github.com/start2020/MedTVL
作者：Jiexia Ye (Data Science and Analytics Thrust, The Hong Kong University of Science and Technology (Guangzhou)); Jia Li (Data Science and Analytics Thrust, The Hong Kong University of Science and Technology (Guangzhou)); Fugee Tsung (Department of Industrial Engineering & Decision Analytics, The Hong Kong University of Science and Technology)
关键词：医学时间序列分类；多模态学习；对比学习
29 From Causal Discovery to Dynamic Causal Inference in Neural Time Series
链接：https://dl.acm.org/doi/10.1145/3770855.3818956
作者：Dmitry Zaytsev (Lucy Family Institute for Data & Society, University of Notre Dame); Valentina V. Kuskova (Lucy Family Institute for Data & Society, University of Notre Dame); Michael Coppedge (Political Science, University of Notre Dame)
关键词：动态因果推断；时变自回归；冲击响应分析；反事实模拟；结构先验
