aiqbench analysis
This research introduces MASHT, a novel pipeline combining Random Convolutional Features (MultiRocket/Hydra) with in-context tabular foundation models to eliminate the need for task-specific training. The method matches or exceeds state-of-the-art baselines like HIVE-COTE 2.0 on both univariate and multivariate time series datasets by leveraging pre-trained representations directly during inference. This approach offers a promising direction for deploying efficient, high-performance classification in critical domains such as medical signal analysis without heavy fine-tuning overheads.
In-Context Time Series Classification with Random Convolutional Features
arXiv:2607.19234v1 Announce Type: new Abstract: Time series classification is central to domains like medical signal analysis, industrial monitoring, and sensor-based activity recognition, where class information manifests as localized shapes, specific frequencies, temporal shifts, or complex cross-channel interactions. Random convolutional transforms efficiently map these sequences to fixed-dimensional tabular features but are traditionally paired with simple linear classifiers. We investigate whether a pretrained tabular foundation model can more effectively harness these rich representations. We propose MASHT, a pipeline that marries MultiRocket and Hydra features with the power of in-context tabular foundation models. By leveraging a pretrained tabular foundation model, our approach completely bypasses task-specific model training, requiring only feature extraction and direct inference. Extensive experiments demonstrate that MASHT matches state-of-the-art time series classification baselines on univariate tasks, achieving a lower average rank than HIVE-COTE 2.0. On multivariate datasets, MASHT remains highly competitive with the strongest reference methods.
Abstract:Time series classification is central to domains like medical signal analysis, industrial monitoring, and sensor-based activity recognition, where class information manifests as localized shapes, specific frequencies, temporal shifts, or complex cross-channel interactions. Random convolutional transforms efficiently map these sequences to fixed-dimensional tabular features but are traditionally paired with simple linear classifiers. We investigate whether a pretrained tabular foundation model can more effectively harness these rich representations. We propose MASHT, a pipeline that marries MultiRocket and Hydra features with the power of in-context tabular foundation models. By leveraging a pretrained tabular foundation model, our approach completely bypasses task-specific model training, requiring only feature extraction and direct inference. Extensive experiments demonstrate that MASHT matches state-of-the-art time series classification baselines on univariate tasks, achieving a lower average rank than HIVE-COTE 2.0. On multivariate datasets, MASHT remains highly competitive with the strongest reference methods.
Machine Learning (cs.LG)
arXiv:2607.19234 [cs.LG]
(or arXiv:2607.19234v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2607.19234 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 234 words
Related coverage
Generating Financial Time Series by Matching Random Convolutional Features
arXiv · Jun 5, 2026
Cracking the Data Science Case Study Interview
Analytics Vidhya · Jul 26, 2026
Links #5: 2026/07
LessWrong · Jul 31, 2026
ChorusTIC: Training-Free Multivariate Time Series Classification via Chorus In-Context Learning
arXiv · Aug 27, 2026