Time Series Generation

Follow this topic Ask about this topic

Latest papers 84

15 papers in the last four weeks, up 114% on the four weeks before. 0.1% of all new papers.

All topics

Cross-View Time-Frequency Diffusion for Extreme-Aware Time-Series Generation
Oct 12, 2026Lin Jiang, Rongchao Xu, Guang WangTime Series Generation Diffusion Models
Department of Computer Science, Florida State University, Tallahassee, FL 32306, USA
IND-Eval: Aligning Evaluation and Simulation for Industrial Time Series Foundation Models
Oct 12, 2026Vebjørn Haugland, Pavel Gulin, David Foss +5Time Series Forecasting Time-Series Foundation Models
Cognite
MontageFlow: Physically Valid High Fidelity EEG Generation
Oct 10, 2026Sudarshan Regmi, Nabaraj Subedi, Arvind Pillai +6Time Series Generation Constrained Generative Modeling
Dartmouth College
Evaluating Sequence Assembly Strategies for Differentially Private Synthetic Time-Series Forecasting
Oct 7, 2026Guoxiong Long, Huizhen Huang, Qikun Cai +2Synthetic Data Evaluation Synthetic Data Generation
School of Computer and Big Data, Minjiang University, No. 200 Xiyuangong Road, Fuzhou, 350108, Fujian, China
Sensor Geometry as a Flow-Matching Prior for Multi-Channel Brain Signals
Oct 6, 2026Jaedong HwangElectroencephalography Time Series Generation
Massachusetts Institute of Technology
HyperNSDE: Personalized Neural SDEs for Joint Static-Longitudinal Clinical Data Generation
Oct 5, 2026Perrine Chassat, Agathe GuillouxStochastic Differential Equations Synthetic Data Generation
Inria, Université Paris Cité, Inserm, HeKA F-75015 Paris, France
Image Synthesis as an Intermediate for Controllable Time Series Generation
Oct 4, 2026Haochen Yuan, Jing Xie, Yunbo WangTime Series Forecasting Time Series Generation
MoE Key Lab of Artificial Intelligence, Institute of AI, School of Computer Science, Shanghai Jiao Tong University, Shanghai, China
Wavelet Flow Matching for Time Series
Sep 30, 2026Lucas Poinsignon, Jorge da Silva Goncalves, Samuel Ruiperez-Campillo +1Flow Matching Time Series Generation
Dept. of Computer Science ETH Zurich, Switzerland
Synthesis Without Training: An Inference-Only Pipeline for Tabular, Temporal, and Relational Synthetic Data
Sep 29, 2026Zilong Zhao, Abdul Raheem, Jiayu Li +5Constrained Generative Modeling Synthetic Data Generation
Betterdata AI · National University of Singapore · University of Illinois Urbana-Champaign +1
HALO: Enhancing Time Series Generation via Hyperspherical Latents and Masked AutoregRessive Modeling
Sep 28, 2026Chunyi Hou, Xiangfei Qiu, Hanyin Cheng +2Variational Autoencoders Time Series Generation
School of Data Science and Engineering, East China Normal University, Shanghai, China
StatD2GAN: When Calibration Masks Generator Quality in Held-Out Evaluation of Synthetic Weather Sequences
Sep 27, 2026Mustafa Ozaytac, Ozge Karadag AtasTime Series Generation Model Calibration
Department of Statistics, Graduate School of Science and Engineering, Hacettepe University, Ankara, 06800, Turkey · Department of Statistics, Faculty of Science, Hacettepe University, Ankara, 06800, Turkey
ChronoFlow: Hierarchical Flow Matching for Irregular Time Series Generation
Sep 27, 2026Changhun Kim, Sunguk Jang, Jeongjun Lee +5Irregular Time-Series Modeling Flow Matching
University of Wisconsin–Madison · AITRICS · KAIST
ShapeLex: Decoupling Local Shape Symbolization and Global Scale Modeling for Text-Controlled Time Series Generation
Sep 21, 2026Subo Wei, Jianqi Gao, Mingyan Fan +3Time Series Generation Autoregressive Generation
School of Computer Engineering and Science, Shanghai University, Shanghai, China · Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai, China
Time series generation with spectrally aligned latent flow matching
Sep 18, 2026Camilo Carvajal Reyes, Felipe TobarTime Series Generation Flow-Based Generative Modeling
Department of Mathematics, Imperial College London, 180 Queen’s Gate, London SW7 2HR, United Kingdom
LLMs as Master Forgers: Generating Synthetic Time Series Data for Manufacturing
Sep 14, 2026Mantek Singh, Jeshwanth Challagundla, Prateek Karnal +3Manufacturing Synthetic Data Generation
Liverpool John Moores University Liverpool, England · University of Texas at Arlington Texas, USA · IIT Patna Patna, India +3
CodeTS: Verifiable Text-to-Time Series Generation via Executable Code
Sep 14, 2026Xudong Yuan, Shunyu Liu, Tongya Zheng +3Time Series Generation
Zhejiang University, Hangzhou, China · Nanyang Technological University, Singapore · South China University of Technology, Guangzhou, China
Representing Clinical Conditions on Vital Signs from Healthy Individuals using Latent Modeling
Sep 14, 2026Rafael Pina, Varuna De Silva, Mindula IlleperumaSynthetic Data Augmentation Time Series Generation
Institute for Digital Technologies Loughborough University London London, United Kingdom
MedFlow: Class-Aware Multi-Scale Generation for Medical Time-Series Synthesis
Sep 4, 2026Yanhao Huang, Shibo Feng, Wanjin Feng +2Flow Matching Synthetic Data Generation
Shanghai Jiao Tong University · Nanyang Technological University · Tsinghua University
EyeMakeYou: Identity-, Task-, and Subjective-State-Conditioned Diffusion for High-Frequency Gaze Synthesis
Sep 3, 2026Kamrul Hasan, Mehedi Hasan Raju, Oleg V. KomogortsevSynthetic Data Augmentation Conditional Generative Modeling
Texas State University, San Marcos, Texas, USA
A Synthetic Multivariate Refrigerator Time-Series Dataset for Predictive Maintenance
Sep 3, 2026Islam Benamirouche, Feriel Fass, Djemel ZiouSynthetic Data Generation Time Series Generation
Département d’informatique, Université de Sherbrooke, Sherbrooke, QC, Canada
SegTSim: A Big Data Driven Segmented Temporal Simulation Framework for Heterogeneous Multivariate Systems
Aug 29, 2026Xinhang Li, Chenxi Geng, Yujia SunMultivariate Time Series Forecasting Time Series Generation
College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China
Conditional Diffusion Models for Energy-Efficient Driving
Aug 28, 2026Hemanth Neelgund Ramesh, André Snoeck, Chyi-Fu Hong +1Conditional Diffusion Models Time Series Generation
Department of Mechanical Engineering, University of Washington · Amazon Science · Department of Materials Science and Metallurgy, University of Cambridge
A Locally Tokenized Generative Model for Robust Time-Series Watermarking
Aug 20, 2026Dongbin Kim, Geonwoo Shin, Yujin Choi +2Robust Watermarking Vector-Quantized Variational Autoencoder
Seoul National University · Nanyang Technological University
CAST: Context- and Anomaly Structure-Conditioned Time Series Anomaly Generation
Aug 19, 2026Haochen Zhang, Jie Peng, Songyuan Sui +3Synthetic Anomaly Generation Time Series Generation
UNITES Lab, University of North Carolina at Chapel Hill · Rice University · Oregon State University
GALA: Generation-Aware Cross-Modal Alignment for Text-to-Time-Series Synthesis
Aug 13, 2026Haochen Zhang, Gengwei Zhang, Laura Yao +2Time Series Generation Conditional Generative Modeling
UNITES Lab, University of North Carolina at Chapel Hill
Represent, Then Generate: Multimodal-Conditioned Time-Series Generation under Irregular Missingness
Aug 12, 2026Haochen Zhang, Jiaheng Guo, Yu-Chao Huang +2Learning with Missing Data Time Series Generation
UNITES Lab, University of North Carolina at Chapel Hill
Physics-informed Diffusion Generative Model for Time-Series Data Synthesis in Dynamic Systems
Aug 11, 2026Haiteng Wang, Yunfei Zhu, Tao Wang +4Synthetic Data Augmentation Synthetic Data Generation
School of Automation Science and Electrical Engineering, Beihang University, Beijing, 100191, China. · School of Software, Beihang University, Beijing, 100191, China. · Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Kowloon, 100191, Hong Kong. +1
Benchmarking Time Series Generation Methods for Privacy-Preserving Forecasting
Aug 11, 2026Luis Amorim, Vitor Cerqueira, Moises Santos +2Synthetic Data Generation Time Series Forecasting
Faculdade de Engenharia da Universidade do Porto, Porto, Portugal · University of Coimbra, Coimbra, Portugal · Laboratory for Artificial Intelligence and Computer Science (LIACC), Portugal +3
Evaluating Generative Time-Series Models on Data with Point Masses
Aug 10, 2026Jian XuTime Series Forecasting Time Series Generation
RIKEN iTHEMS, Wako, Japan · RIKEN Center for Advanced Intelligence Project (AIP), Tokyo, Japan
Omni-modal decomposition autoencoders learn full-stack wearable disentangled representations
Aug 7, 2026Ioannis Ziogas, Ensieh Khazaei, Bilal Taha +4Variational Autoencoders Human Activity Recognition
Department of Biomedical Engineering and Biotechnology, Khalifa University, 127788 Abu Dhabi, UAE · Edward S. Rogers Sr. Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON M5S 1A1, Canada · MIT Media Lab +2

Oct 12, 2026J/K move · Enter open · S save

Cross-View Time-Frequency Diffusion for Extreme-Aware Time-Series Generation

Lin Jiang, Rongchao Xu, Guang Wang

Department of Computer Science, Florida State University, Tallahassee, FL 32306, USA

Time Series Generation Diffusion Models

Open paper AskPDF arXiv

Generating realistic time-series requires faithfully reproducing rare but high-impact extreme events. Existing approaches often rely on specialized representations or extreme-focused training objectives, which may restrict the diversity of captured extremes or compromise overall distributional fidelity. We propose Cross-View Time-Frequency Diffusion (CrossTF), a framework that exploits discrepancies between time- and frequency-view denoisers for extreme-aware time-series generation. Our key empirical finding is extreme-sensitive cross-view asymmetry: moving along the direction connecting the two views' clean-sample estimates substantially changes extreme fidelity while having limited effects on overall generation quality. We further identify a late-stage cross-view window in which this discrepancy stabilizes and cross-view adjustments have a particularly strong effect on extreme fidelity. These findings motivate a staged optimization strategy in CrossTF: after training and freezing both denoisers, it only needs to learn step-specific fusion parameters within this window using an Extreme Value Theory-based objective under an energy-distance constraint. This formulation supports interpolation between and extrapolation beyond the view-specific estimates while limiting degradation in overall generation quality. Extensive experiments on seven real-world datasets against 11 baselines show that CrossTF improves extreme fidelity by 26.1% and overall performance by 7.4% relative to the strongest baselines. These results highlight cross-view discrepancy as an effective basis for low-dimensional optimization in extreme-aware time-series generation.