The study evaluates how much self‑supervised pretraining contributes to time series foundation models (TSFMs) by comparing seven publicly released TSFMs with architecture‑matched randomly initialized versions on the UEA multivariate time series classification archive. Across 183 dataset‑level comparisons, pretraining yields modest accuracy gains (1.43–5.19 percentage points, average 2.58 points), with only one model (NuTime) remaining significant after Holm correction. The results highlight that strong non‑pretrained baselines such as ROCKET, Hydra, and Quant can compete without large‑scale pretraining, underscoring the need to separate architectural effects from pretrained weights in MTSC evaluations.
arXiv:2611.06988v1 Announce Type: cross Abstract: Forecasting foundation models are increasingly used as frozen feature extractors for time series classification, but it remains unclear whether their...
arXiv:2607. 00958v1 Announce Type: new Abstract: Time series are central to modern data mining applications, from industrial telemetry and server metrics to finance and physiology, yet time-series self-supervised learning often depends on view and augmentation choices that encode domain-specific invariances.
arXiv:2607. 04919v1 Announce Type: new Abstract: Deploying a time series foundation model requires GPU infrastructure, engineering overhead, and carries no guarantee of improvement over XGBoost.
arXiv:2609.39386v1 Announce Type: new Abstract: Pretrained time-series foundation models (TSFMs) are evaluated as forecasters of future values, yet for sparse series many decisions depend only on whi...
Self-supervised pretraining has produced capable time-series foundation models, but its value for Prognostics and Health Management (PHM) has not been measured under leakage-controlled conditions. We...
arXiv:2609.37255v1 Announce Type: cross Abstract: Time series foundation models (TSFMs) are pretrained on heterogeneous collections containing billions of observations, yet their training windows are...
The study evaluates self‑supervised pretraining for Prognostics and Health Management (PHM) across 47 public datasets, using a compact Transformer encoder trained with masked patch prediction. Transfer performance is highly task‑dependent: it dramatically improves fault‑classification on CWRU and HSG18 but offers no reliable benefit for remaining‑useful‑life estimation on CMAPSS. The authors also compare centralized versus simulated federated pretraining, finding that centralized training generally outperforms federated training under matched budgets, and that the intrinsic SSL loss does not reliably predict downstream performance.
TIGER is a time‑series classification method that uses a small set of three generic classifiers applied to four different representation families, producing twelve base learners. The predictions are stacked into a meta‑feature matrix and an adaptive meta‑classifier—choosing between a weighted hard majority vote and the pretrained TabICLv2 model—selects the best rule per dataset based on training sample size. On a 142‑dataset UCR benchmark, TIGER achieves the highest mean accuracy, balanced accuracy, and F1‑score among six compared algorithms, outperforming each constituent method and demonstrating strong generalization with a single hyperparameter.
arXiv:2607. 07500v1 Announce Type: cross Abstract: Time series classification (TSC) is dominated by a two-stage paradigm: train a feature encoder -- either from scratch on the target dataset or via pretraining on large corpora -- and then fit a task-specific classifier on top.
arXiv:2606. 03938v1 Announce Type: cross Abstract: Multi-epoch training is becoming the standard now that compute is growing faster than the supply of high-quality text.
arXiv:2606. 08578v1 Announce Type: new Abstract: Recently, large time series models (LTSMs) have gained increasing attention due to their similarities to large language models, including flexible context length, scalability, and task generality, outperforming advanced task-specific models.
The paper introduces the Recursive Time-Series Model (RTSM), a deep-learning architecture that uses a learned convolutional temporal tokenizer, a single-state weight‑tied refinement module, and a pooled classification readout to classify multivariate time‑series data. On the 30 UEA datasets, RTSM achieves a mean accuracy of 72.81% under validation‑based configuration, surpassing all 17 evaluated deep‑learning models and outperforming MambaSL by 1.60 percentage points. Ablation studies show that the proposed tokenizer and single‑state design outperform simpler convolutional embeddings and two‑state variants, and the full classifier reaches 86.10% mean accuracy compared to 84.79% for a tokenizer‑only classifier.