DiTS: Multimodal Diffusion Transformers Are Time Series Forecasters

• URL: http://arxiv.org/abs/2602.06597v1
• Date: Fri, 06 Feb 2026 10:48:13 GMT
• Title: DiTS: Multimodal Diffusion Transformers Are Time Series Forecasters
• Authors: Haoran Zhang, Haixuan Liu, Yong Liu, Yunzhong Qiu, Yuxuan Wang, Jianmin Wang, Mingsheng Long,
• Abstract summary: Existing generative time series models do not address the multi-dimensional properties of time series data well.<n>Inspired by Multimodal Diffusion Transformers that integrate textual guidance into video generation, we propose Diffusion Transformers for Time Series (DiTS)
• Score: 50.43534351968113
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: While generative modeling on time series facilitates more capable and flexible probabilistic forecasting, existing generative time series models do not address the multi-dimensional properties of time series data well. The prevalent architecture of Diffusion Transformers (DiT), which relies on simplistic conditioning controls and a single-stream Transformer backbone, tends to underutilize cross-variate dependencies in covariate-aware forecasting. Inspired by Multimodal Diffusion Transformers that integrate textual guidance into video generation, we propose Diffusion Transformers for Time Series (DiTS), a general-purpose architecture that frames endogenous and exogenous variates as distinct modalities. To better capture both inter-variate and intra-variate dependencies, we design a dual-stream Transformer block tailored for time-series data, comprising a Time Attention module for autoregressive modeling along the temporal dimension and a Variate Attention module for cross-variate modeling. Unlike the common approach for images, which flattens 2D token grids into 1D sequences, our design leverages the low-rank property inherent in multivariate dependencies, thereby reducing computational costs. Experiments show that DiTS achieves state-of-the-art performance across benchmarks, regardless of the presence of future exogenous variate observations, demonstrating unique generative forecasting strengths over traditional deterministic deep forecasting models.

Related papers

• Gateformer: Advancing Multivariate Time Series Forecasting through Temporal and Variate-Wise Attention with Gated Representations [2.4302562182247636]
  We re-purpose the Transformer architecture to model both cross-time and cross-variate dependencies.<n>Our method achieves state-of-the-art performance across 13 real-world datasets, delivering performance improvements up to 20.7% over original models.
  arXiv  Detail & Related papers  (2025-05-01T04:59:05Z)
• Timer-XL: Long-Context Transformers for Unified Time Series Forecasting [67.83502953961505]
  We present Timer-XL, a causal Transformer for unified time series forecasting.<n>Based on large-scale pre-training, Timer-XL achieves state-of-the-art zero-shot performance.
  arXiv  Detail & Related papers  (2024-10-07T07:27:39Z)
• TiVaT: A Transformer with a Single Unified Mechanism for Capturing Asynchronous Dependencies in Multivariate Time Series Forecasting [4.733959271565453]
  TiVaT is a novel architecture incorporating a single unified module, a Joint-Axis (JA) attention module.<n>The JA attention module dynamically selects relevant features to particularly capture asynchronous interactions.<n>Extensive experiments demonstrate TiVaT's overall performance across diverse datasets.
  arXiv  Detail & Related papers  (2024-10-02T13:24:24Z)
• 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)
• PRformer: Pyramidal Recurrent Transformer for Multivariate Time Series Forecasting [82.03373838627606]
  Self-attention mechanism in Transformer architecture requires positional embeddings to encode temporal order in time series prediction. We argue that this reliance on positional embeddings restricts the Transformer's ability to effectively represent temporal sequences. We present a model integrating PRE with a standard Transformer encoder, demonstrating state-of-the-art performance on various real-world datasets.
  arXiv  Detail & Related papers  (2024-08-20T01:56:07Z)
• 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)
• Unified Training of Universal Time Series Forecasting Transformers [104.56318980466742]
  We present a Masked-based Universal Time Series Forecasting Transformer (Moirai) Moirai is trained on our newly introduced Large-scale Open Time Series Archive (LOTSA) featuring over 27B observations across nine domains. Moirai achieves competitive or superior performance as a zero-shot forecaster when compared to full-shot models.
  arXiv  Detail & Related papers  (2024-02-04T20:00:45Z)
• Towards Long-Term Time-Series Forecasting: Feature, Pattern, and Distribution [57.71199089609161]
  Long-term time-series forecasting (LTTF) has become a pressing demand in many applications, such as wind power supply planning. Transformer models have been adopted to deliver high prediction capacity because of the high computational self-attention mechanism. We propose an efficient Transformerbased model, named Conformer, which differentiates itself from existing methods for LTTF in three aspects.
  arXiv  Detail & Related papers  (2023-01-05T13:59:29Z)
• W-Transformers : A Wavelet-based Transformer Framework for Univariate Time Series Forecasting [7.075125892721573]
  We build a transformer model for non-stationary time series using wavelet-based transformer encoder architecture. We evaluate our framework on several publicly available benchmark time series datasets from various domains.
  arXiv  Detail & Related papers  (2022-09-08T17:39:38Z)
• 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.