ACM MM 2025于2025年10月27号-31号在爱尔兰都柏林（Dublin, Ireland）举行

本文总结了ACM MM 2025有关时间序列（Time Series）相关文章。如有疏漏，欢迎大家补充。

时间序列Topic：预测，分类，异常检测，医疗时序等

1. TimesBERT: A BERT-Style Foundation Model for Time Series Understanding 2. DualSG: A Dual-Stream Explicit Semantic-Guided Multivariate Time Series Forecasting Framework 3. Towards Measuring and Modeling Geometric Structures in Time Series Forecasting via Image Modality 4. DisMS-TS: Eliminating Redundant Multi-scale Features for Time Series Classification 5. Reading Between the Channels: Knowledge-Augmented Medical Time Series Classification 6. Foresail: LLM Sensor Knowledge Empowered Status-guided Network for Multivariate Time-series Classification 7. Unsupervised Dual-Domain Memory Model for Time Series Anomaly Detection 8. RSFomer: Time Series Transformer for Robust Sports Action Recognition 9. A Comprehensive Benchmark for Electrocardiogram Time-Series

1 TimesBERT: A BERT-Style Foundation Model for Time Series Understanding

链接：https://dl.acm.org/doi/10.1145/3746027.3755238

作者：Haoran Zhang, Yong Liu, Yunzhong Qiu, Haixuan Liu, Zhongyi Pei, Jianmin Wang, Mingsheng Long

关键词：基础模型，预训练，Bert

2 DualSG: A Dual-Stream Explicit Semantic-Guided Multivariate Time Series Forecasting Framework

链接：https://dl.acm.org/doi/10.1145/3746027.3755458

代码：https://github.com/BenchCouncil/DualSG

作者：Kuiye Ding, Fanda Fan, Yao Wang, Ruijie Jian, Xiaorui Wang, Luqi Gong, Yishan Jiang, Chunjie Luo, Jianfeng Zhan

关键词：预测，大模型

3 Towards Measuring and Modeling Geometric Structures in Time Series Forecasting via Image Modality

链接：https://dl.acm.org/doi/10.1145/3746027.3754794

作者：Mingyang Yu, Xiahui Guo, Peng Chen, Zhenkai Li, Yang Shu

关键词：预测，几何结构

4 DisMS-TS: Eliminating Redundant Multi-scale Features for Time Series Classification

链接：https://dl.acm.org/doi/10.1145/3746027.3754842

代码：https://github.com/ZhipengLiu75/DisMS-TS

作者：Zhipeng Liu, Peibo Duan, Binwu Wang, Xuan Tang, Qi Chu, Changsheng Zhang, Yongsheng Huang, Bin Zhang

关键词：分类，特征解耦

5  Reading Between the Channels: Knowledge-Augmented Medical Time Series Classification

链接：https://dl.acm.org/doi/10.1145/3746027.3755824

作者：Xiaoyan Yuan, Wei Wang, Junxin Chen, Xiping Hu

关键词：医疗时序，医疗文本知识，通道相关性，预训练语言模型

6 Foresail: LLM Sensor Knowledge Empowered Status-guided Network for Multivariate Time-series Classification

链接：https://dl.acm.org/doi/10.1145/3746027.3755806

代码：https://github.com/weatherjyh/Foresail

作者：Yuhan Jing, Bo He, Haifeng Sun, Qi Qi, Zirui Zhuang, Lei Zhang, Jianxin Liao, Jing-Yu Wang

关键词：分类，大模型，知识抽取

7 Unsupervised Dual-Domain Memory Model for Time Series Anomaly Detection

链接：https://dl.acm.org/doi/10.1145/3746027.3754892

作者：Mingle Zhou, Xingli Wang, Jiachen Li, Delong Han, Gang Li

关键词：异常检测，无监督

8 A Comprehensive Benchmark for Electrocardiogram Time-Series

链接：https://dl.acm.org/doi/10.1145/3746027.3754729

代码：https://github.com/ZhijiangTang/ECG-Benchmark

作者：Zhijiang Tang, Jiaxin Qi, Yuhua Zheng, Jianqiang Huang

关键词：心电图，benchmark

9 RSFomer: Time Series Transformer for Robust Sports Action Recognition

链接：https://dl.acm.org/doi/10.1145/3746027.3754807

代码：https://github.com/Limber0117/RSFomer

作者：Yongan Guo, Zhongyan Zhou, Yuao Wang, Na Zhu, Xuyun Zhang, Hongwang Xiao, Yuan Miao, Bo Li

关键词：人体活动识别