TSGym: Design Choices for Deep Multivariate Time-Series Forecasting

• URL: http://arxiv.org/abs/2509.17063v1
• Date: Sun, 21 Sep 2025 12:49:31 GMT
• Title: TSGym: Design Choices for Deep Multivariate Time-Series Forecasting
• Authors: Shuang Liang, Chaochuan Hou, Xu Yao, Shiping Wang, Minqi Jiang, Songqiao Han, Hailiang Huang,
• Abstract summary: This work bridges gaps by decomposing deep MTSF methods into their core, fine-grained components.<n>We propose a novel automated solution called TSGym for MTSF tasks.<n>Extensive experiments indicate that TSGym significantly outperforms existing state-of-the-art MTSF and AutoML methods.
• Score: 38.12202305030755
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recently, deep learning has driven significant advancements in multivariate time series forecasting (MTSF) tasks. However, much of the current research in MTSF tends to evaluate models from a holistic perspective, which obscures the individual contributions and leaves critical issues unaddressed. Adhering to the current modeling paradigms, this work bridges these gaps by systematically decomposing deep MTSF methods into their core, fine-grained components like series-patching tokenization, channel-independent strategy, attention modules, or even Large Language Models and Time-series Foundation Models. Through extensive experiments and component-level analysis, our work offers more profound insights than previous benchmarks that typically discuss models as a whole. Furthermore, we propose a novel automated solution called TSGym for MTSF tasks. Unlike traditional hyperparameter tuning, neural architecture searching or fixed model selection, TSGym performs fine-grained component selection and automated model construction, which enables the creation of more effective solutions tailored to diverse time series data, therefore enhancing model transferability across different data sources and robustness against distribution shifts. Extensive experiments indicate that TSGym significantly outperforms existing state-of-the-art MTSF and AutoML methods. All code is publicly available on https://github.com/SUFE-AILAB/TSGym.

Related papers

• TokaMind: A Multi-Modal Transformer Foundation Model for Tokamak Plasma Dynamics [56.073642366268764]
  TokaMind is an open-source foundation model framework for fusion plasma modeling.<n>It is trained on heterogeneous tokamak diagnostics from the publicly available MAST dataset.<n>We evaluate TokaMind on the recently introduced MAST benchmark TokaMark.
  arXiv  Detail & Related papers  (2026-02-16T12:26:07Z)
• EMTSF:Extraordinary Mixture of SOTA Models for Time Series Forecasting [0.750638869146118]
  We propose a strong Mixture of Experts (MoE) framework for Time Series Forecasting.<n>Our method combines the state-of-the-art (SOTA) models including xLSTM, en hanced Linear, PatchTST, and minGRU.<n>Our proposed model outperforms all existing TSF models on standard benchmarks, surpassing even the latest approaches based on MoE frameworks.
  arXiv  Detail & Related papers  (2025-10-27T14:55:30Z)
• MOMEMTO: Patch-based Memory Gate Model in Time Series Foundation Model [0.07777489763207261]
  We propose textbfMOMEMTO, a TFM for anomaly detection enhanced with a patch-based memory module.<n>The memory module is designed to capture representative normal patterns from multiple domains and enables a single model to be jointly fine-tuned.<n> Experimental results demonstrate that MOMEMTO, as a single model, achieves higher scores on AUC and VUS metrics compared to baseline methods.
  arXiv  Detail & Related papers  (2025-09-23T07:48:25Z)
• Multi-Scale Finetuning for Encoder-based Time Series Foundation Models [56.503053716053]
  Time series foundation models (TSFMs) demonstrate impressive zero-shot performance for time series forecasting.<n>We argue that it falls short of fully leveraging TSFMs' capabilities, often resulting in overfitting and suboptimal performance.<n>We propose textbftextscfinetextbftextsctuning (textbfMSFT), a simple yet general framework that explicitly integrates multi-scale modeling into the finetuning process.
  arXiv  Detail & Related papers  (2025-06-17T01:06:01Z)
• TSFM-Bench: A Comprehensive and Unified Benchmark of Foundation Models for Time Series Forecasting [35.505530132151]
  Time Series Forecasting (TSF) is key functionality in numerous fields, such as financial investment, weather services, and energy management.<n>Many TSF methods require domain-specific data collection and model training and do not generalize well when applied in other domains.<n>Time Series Foundation Models (TSFMs) that are pre-trained on massive heterogeneous time series data aim to overcome these limitations.
  arXiv  Detail & Related papers  (2024-10-15T17:23:49Z)
• TimeDiT: General-purpose Diffusion Transformers for Time Series Foundation Model [11.281386703572842]
  TimeDiT is a diffusion transformer model that combines temporal dependency learning with probabilistic sampling.<n>TimeDiT employs a unified masking mechanism to harmonize the training and inference process across diverse tasks.<n>Our systematic evaluation demonstrates TimeDiT's effectiveness both in fundamental tasks, i.e., forecasting and imputation, through zero-shot/fine-tuning.
  arXiv  Detail & Related papers  (2024-09-03T22:31:57Z)
• LTSM-Bundle: A Toolbox and Benchmark on Large Language Models for Time Series Forecasting [69.33802286580786]
  We introduce LTSM-Bundle, a comprehensive toolbox, and benchmark for training LTSMs.<n>It modularized and benchmarked LTSMs from multiple dimensions, encompassing prompting strategies, tokenization approaches, base model selection, data quantity, and dataset diversity.<n> Empirical results demonstrate that this combination achieves superior zero-shot and few-shot performances compared to state-of-the-art LTSMs and traditional TSF methods.
  arXiv  Detail & Related papers  (2024-06-20T07:09:19Z)
• UniTST: Effectively Modeling Inter-Series and Intra-Series Dependencies for Multivariate Time Series Forecasting [98.12558945781693]
  We propose a transformer-based model UniTST containing a unified attention mechanism on the flattened patch tokens. Although our proposed model employs a simple architecture, it offers compelling performance as shown in our experiments on several datasets for time series forecasting.
  arXiv  Detail & Related papers  (2024-06-07T14:39:28Z)
• ConvTimeNet: A Deep Hierarchical Fully Convolutional Model for Multivariate Time Series Analysis [7.979501926410114]
  ConvTimeNet is a hierarchical pure convolutional model designed for time series analysis.<n>It adaptively perceives local patterns of temporally dependent basic units in a data-driven manner.<n>A large kernel mechanism is employed to ensure that convolutional blocks can be deeply stacked.
  arXiv  Detail & Related papers  (2024-03-03T12:05:49Z)
• Parsimony or Capability? Decomposition Delivers Both in Long-term Time Series Forecasting [46.63798583414426]
  Long-term time series forecasting (LTSF) represents a critical frontier in time series analysis. Our study demonstrates, through both analytical and empirical evidence, that decomposition is key to containing excessive model inflation. Remarkably, by tailoring decomposition to the intrinsic dynamics of time series data, our proposed model outperforms existing benchmarks.
  arXiv  Detail & Related papers  (2024-01-22T13:15:40Z)
• Multi-scale Attention Flow for Probabilistic Time Series Forecasting [68.20798558048678]
  We propose a novel non-autoregressive deep learning model, called Multi-scale Attention Normalizing Flow(MANF) Our model avoids the influence of cumulative error and does not increase the time complexity. Our model achieves state-of-the-art performance on many popular multivariate datasets.
  arXiv  Detail & Related papers  (2022-05-16T07:53:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.