Time Series Tokenization

Latest papers

15 papers3 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Generative Modeling for Cascading Extremes via Autoregressive Transformers
Nov 3, 2026Miguel de Carvalho, Myung Won Lee, Jordan Richards +1Generative Modeling Time Series Tokenization
School of Mathematics and Maxwell Institute for Mathematical Sciences University of Edinburgh · CIDMA–Department of Mathematics Universidade de Aveiro · National Technical University of Athens Athena/Archimedes Research Centre
TEMPO: Time Series Understanding via Discrete Tokenization
Nov 3, 2026Riccardo MaggioniTime Series Tokenization Time Series Language Modeling
Forgis
EMG Tokenization Preserves Movement Information at a Fraction of the Bitrate
Oct 23, 2026Raul C. Sîmpetru, Amin Olamazadeh, Devon Rohlf +2Time Series Tokenization Surface Electromyography
Department of Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, 91052 Erlangen, Germany
Rethinking Time Series Tokenization from a Frequency Perspective
Oct 20, 2026Zelong Tian, Ming Jin, Bo Du +1Time Series Tokenization Frequency-Domain Time Series Modeling
Griffith University
Trajectory-Guided Tokenization of Complex CSI for Wi-Fi Sensing
Oct 5, 2026Ziyi Wang, Kenuo Xu, Jichu Jiang +8Gesture Recognition Wi-Fi Sensing
Technische Universität Berlin, Berlin, Germany · Huawei Heisenberg Research Center (Munich) · Huawei Technologies Co., Ltd., Shanghai, China +1
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
STELLA: Efficient Sensor-to-LLM Translation for On-Device Human Activity Recognition
Jul 3, 2026Nirhoshan Sivaroopan, Albert Zomaya, Kanchana ThilakarathnaOn-Device Language Model Inference Human Activity Recognition
The University of Sydney · Australia
Time Series as Language: A Universal Tokenizer for General-Purpose Time Series Foundation Models
May 31, 2026Yunhao Zhang, Ruiying Qi, Jiale Zheng +3Time Series Tokenization Time Series Generation
Shanghai Jiao Tong University · Huawei Noah’s Ark Lab
Continuity and Ordinality Matter: Constraining Time Series Tokens for Effective Time Series Analysis with Large Language Models
May 22, 2026Musheng Li, Ziying Zhang, Cheng jin +1Time Series Forecasting Time Series Representation Learning
Department of Electronic Engineering Tsinghua University
DeepTokenEEG Enhancing Mild Cognitive Impairment and Alzheimers Classification via Tokenized EEG Features
May 14, 2026Thinh Nguyen-Quang, Minh Long Ngo, Ngoc-Son Nguyen +7Electroencephalography Alzheimer's Disease Classification
Hanoi University of Science and Technology, Hanoi, Vietnam · VNU University of Engineering and Technology, Hanoi, Vietnam · Central Military Hospital 108, Hanoi, Vietnam +1
Dywave: Event-Aligned Dynamic Tokenization for Heterogeneous IoT Sensing Signals
May 13, 2026Tomoyoshi Kimura, Denizhan Kara, Jinyang Li +6Internet of Things Time Series Tokenization
University of Illinois Urbana-Champaign · Hong Kong University of Science and Technology · Shanghai Jiao Tong University
TimeTok: Granularity-Controllable Time-Series Generation via Hierarchical Tokenization
May 2, 2026Seokhyun Lee, Jaeho Kim, Changjun Oh +2Time Series Generation Multiscale Time-Series Modeling
Department of Artificial Intelligence Korea University Seoul, South Korea · Department of Applied Mathematics and Theoretical Physics University of Cambridge Cambridge, United Kingdom
A Systematic Evaluation of Sample-Level Tokenization Strategies for MEG Foundation Models
Feb 18, 2026SungJun Cho, Chetan Gohil, Rukuang Huang +2Time Series Tokenization Time-Series Foundation Models
Oxford Centre for Human Brain Activity, University of Oxford, Oxford OX3 7JX, U.K. · Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford OX3 9DU, U.K. · Department of Psychiatry, University of Oxford, Oxford OX3 7JX, U.K. +1
Patch-Level Tokenization with CNN Encoders and Attention for Improved Transformer Time-Series Forecasting
Jan 18, 2026Saurish Nagrath, Saroj Kumar Panigrahy, Soubhagya Sankar BarpandaMultivariate Time Series Forecasting Time Series Transformers
School of Computer Science and Engineering, VIT-AP University, Amaravati, Andhra Pradesh, 522241, India
Mantis: Lightweight Foundation Model for Time Series Classification
Feb 21, 2025Vasilii Feofanov, Songkang Wen, Shifeng Xie +10Time Series Classification Self-Supervised Time Series Representation Learning
1. Huawei Noah’s Ark Lab · 3. Paris Cit´e University · 4. Helmholtz Munich +3

Nov 3, 2026J/K move · Enter open · S save

Generative Modeling for Cascading Extremes via Autoregressive Transformers

Miguel de Carvalho, Myung Won Lee, Jordan Richards, Sotirios Sabanis

School of Mathematics and Maxwell Institute for Mathematical Sciences University of Edinburgh · CIDMA–Department of Mathematics Universidade de Aveiro · National Technical University of Athens Athena/Archimedes Research Centre

Generative Modeling Time Series Tokenization Extreme Value Theory Autoregressive Transformer

Open paper AskPDF arXiv

Extreme events often occur sequentially, from earthquake-tsunami cascades to successive extreme losses in financial markets. We introduce an extremal token representation that encodes multivariate observations as a categorical sequence. A quiet token records the absence of threshold exceedances; otherwise, the token identifies the exceeding component with the largest marginal probability-scale score. The representation is invariant to componentwise strictly increasing transformations. It preserves the identity and temporal ordering of dominant exceedances but discards their magnitudes and the identities of nondominant simultaneous exceedances. A decoder-only transformer estimates the history-dependent next-token kernel on this representation, and recursive sampling generates future token sequences from the induced joint predictive law. We amortize our generative model by pretraining across simulated process families with varying parameters and dependence persistence. Next-token predictions are evaluated against exact conditional probabilities for held-out categorical processes and particle filter references for heavy-tailed vector autoregressive processes. Applied to ten-second returns for five semiconductor stocks, the pretrained model supports zero-shot transfer, while a few epochs of adaptation improve agreement between observed and generated token frequencies and transition patterns. These results support task-specific tokenization and simulation-based pretraining for modeling which dimensions dominate multivariate extremes.