AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting

• URL: http://arxiv.org/abs/2605.25166v1
• Date: Sun, 24 May 2026 16:52:48 GMT
• Title: AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting
• Abstract summary: We propose a structure-guided sparse time series foundation model that aligns expert routing with interpretable temporal structure.<n>AME-TS delivers a strong accuracy-efficiency tradeoff across model scales.<n>It learns more interpretable routing geometry and substantially more stable expert specialization than standard MoE.
• Score: 4.62087302563841
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Time series forecasting models are increasingly scaled through large Transformer backbones, yet most existing approaches process all series through a shared dense computation path despite substantial heterogeneity in temporal structure. Mixture-of-Experts (MoE) offers a natural alternative by enabling conditional computation, but standard MoE routing leaves expert specialization weakly identified and often unstable during downstream adaptation. We propose AME-TS, a structure-guided sparse time series foundation model that aligns expert routing with interpretable temporal structure. AME-TS first uses a lightweight regime predictor to estimate series-level descriptors, including forecastability, seasonality, trend, and sparsity, and maps them to a soft structural prior over experts. This series-level prior guides token-level routing during training, encouraging structure-aligned specialization. On the GIFT-Eval benchmark, AME-TS delivers a strong accuracy-efficiency tradeoff across model scales: it substantially outperforms existing time series foundation models at small model scales and remains competitive with the strongest models at larger scales, while activating substantially fewer parameters through sparse routing. We further show that AME-TS learns more interpretable routing geometry and substantially more stable expert specialization than standard MoE during fine-tuning on the M5 dataset. These results suggest that structure-aware routing is an effective and reliable way to realize the benefits of sparse expert models for time series forecasting.

Related papers

• TS-ICL: A Flexible Time-Indexed Foundation Model for Time Series via In-Context Learning [0.764671395172401]
  Current approaches primarily focus on forecasting, while real-world time series are often irregularly and partially observed.<n>We introduce TS-ICL, a novel probabilistic In-Context Learning encoder--regressor Transformer that unifies forecasting and imputation.
  arXiv  Detail & Related papers  (2026-06-04T08:52:21Z)
• Three-Stage Learning Unlocks Strong Performance in Simple Models for Long-Term Time Series Forecasting [6.603558172147427]
  STAIR aims to unlock the capacity of simple temporal mapping models without introducing complex architectural modules.<n>It first learns common temporal dynamics across variables through a shared temporal mapping, then adapts the shared model to each variable via channel-wise fine-tuning to capture variable-specific patterns, and finally complements the backbone with cross-variable information through residual learning.
  arXiv  Detail & Related papers  (2026-05-13T15:36:46Z)
• Seeking SOTA: Time-Series Forecasting Must Adopt Taxonomy-Specific Evaluation to Dispel Illusory Gains [17.650604674194984]
  We argue that the current practice of evaluating AI/ML time-series forecasting models obscures real progress by overlooking the performance of efficient classical methods.<n>We demonstrate that these "standard" datasets often exhibit dominant autocorrelation patterns and seasonal cycles that can be effectively captured by simpler linear or statistical models.
  arXiv  Detail & Related papers  (2026-03-16T16:34:26Z)
• Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling [51.78972657142583]
  We introduce Timer-S1, a strong Mixture-of-Experts (MoE) time series foundation model with 8.3B total parameters, 0.75B activated parameters for each token, and a context length of 11.5K.<n>To overcome the scalability bottleneck in existing pre-trained time series foundation models, we perform Serial Scaling in three dimensions.
  arXiv  Detail & Related papers  (2026-03-05T04:13:57Z)
• Seg-MoE: Multi-Resolution Segment-wise Mixture-of-Experts for Time Series Forecasting Transformers [0.9058414988965365]
  We introduce Seg-MoE, a sparse MoE design that processes contiguous time-step segments rather than making independent expert decisions.<n>Seg-MoE consistently achieves state-of-the-art forecasting accuracy across almost all prediction horizons.<n>Our results show that aligning the MoE routing granularity with the inherent structure of time series provides a powerful, yet previously underexplored, inductive bias.
  arXiv  Detail & Related papers  (2026-01-29T12:43:35Z)
• Time Series Foundation Models for Process Model Forecasting [8.339024524110828]
  Process Model Forecasting aims to predict how the control-flow structure of a process evolves over time.<n>Machine learning and deep learning models provide only modest gains over statistical baselines.<n>We investigate Time Series Foundation Models (TSFMs) as an alternative for PMF.
  arXiv  Detail & Related papers  (2025-12-08T15:08:50Z)
• A Unified Frequency Domain Decomposition Framework for Interpretable and Robust Time Series Forecasting [81.73338008264115]
  Current approaches for time series forecasting, whether in the time or frequency domain, predominantly use deep learning models based on linear layers or transformers.<n>We propose FIRE, a unified frequency domain decomposition framework that provides a mathematical abstraction for diverse types of time series.<n>Fire consistently outperforms state-of-the-art models on long-term forecasting benchmarks.
  arXiv  Detail & Related papers  (2025-10-11T09:59:25Z)
• Estimating Time Series Foundation Model Transferability via In-Context Learning [74.65355820906355]
  Time series foundation models (TSFMs) offer strong zero-shot forecasting via large-scale pre-training.<n>Fine-tuning remains critical for boosting performance in domains with limited public data.<n>We introduce TimeTic, a transferability estimation framework that recasts model selection as an in-context-learning problem.
  arXiv  Detail & Related papers  (2025-09-28T07:07:13Z)
• Learning Pattern-Specific Experts for Time Series Forecasting Under Patch-level Distribution Shift [51.01356105618118]
  Time series often exhibit complex non-uniform distribution with varying patterns across segments, such as season, operating condition, or semantic meaning.<n>Existing approaches, which typically train a single model to capture all these diverse patterns, often struggle with the pattern drifts between patches.<n>We propose TFPS, a novel architecture that leverages pattern-specific experts for more accurate and adaptable time series forecasting.
  arXiv  Detail & Related papers  (2024-10-13T13:35:29Z)
• Test Time Learning for Time Series Forecasting [1.4605709124065924]
  Test-Time Training (TTT) modules consistently outperform state-of-the-art models, including the Mamba-based TimeMachine.<n>Our results show significant improvements in Mean Squared Error (MSE) and Mean Absolute Error (MAE)<n>This work sets a new benchmark for time-series forecasting and lays the groundwork for future research in scalable, high-performance forecasting models.
  arXiv  Detail & Related papers  (2024-09-21T04:40:08Z)
• 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)
• Timer: Generative Pre-trained Transformers Are Large Time Series Models [83.03091523806668]
  This paper aims at the early development of large time series models (LTSM) During pre-training, we curate large-scale datasets with up to 1 billion time points. To meet diverse application needs, we convert forecasting, imputation, and anomaly detection of time series into a unified generative task.
  arXiv  Detail & Related papers  (2024-02-04T06:55:55Z)
• 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)

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.