arXiv Machine Learning October 12, 2026 By Max Bourgeat

How Much Computation Does Every Forecast Need? Structured Block Skipping in Two Time-Series Foundation Models

Read the original on arXiv Machine Learning →

The paper investigates whether entire Transformer blocks can be skipped during inference in two time‑series foundation models, Chronos‑Bolt and TimesFM 2.5. By applying block‑level masks selected on calibration windows, the authors show that moderate sparsity (e.g., 7/11 blocks for Chronos‑Bolt and 16/20 for TimesFM) preserves forecasting quality while reducing latency by roughly 11–13%. However, more aggressive skipping degrades performance on certain datasets, and dynamic, input‑dependent routing remains unproven across architectures.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv AI

Jun 4

Adaptive Patching Is Harder Than It Looks For Time-Series Forecasting

arXiv:2606. 04074v1 Announce Type: cross Abstract: Adaptive patching is a recent and compelling proposal for time-series Transformers: allocate finer patches where the sequence looks locally informative.

By Federico Zucchi, Yi Xie, Chao Zhang, Keyuan Luo, Thomas Lampert, Ziyue Li

llms benchmarks safety

arXiv Machine Learning

Oct 7

Systematic Evaluation of TabPFN-TS and Chronos-2 for Zero-Shot Heat Load Forecasting in District Heating Networks

arXiv:2608.20024v2 Announce Type: replace Abstract: District heating energy hubs require reliable heat load forecasts for efficient operational scheduling. Forecasting models trained on historical da...

By Ben Spoek, Karim K. Ben Hicham, Kai Derzsi, Philipp Althaus, Alexander Mitsos, Dirk M\"uller

benchmarks

Hugging Face Trending Papers

Jun 25

How Good Can Linear Models Be for Time-Series Forecasting?

Time-series forecasting research has been moving steadily toward larger architectures, from specialized transformers to general-purpose foundation models, on the assumption that capacity is what unlocks accuracy. We take the opposite position: most of the gap can be closed at far lower cost by tuning preprocessing rather than scaling models.

llms benchmarks

arXiv Machine Learning

Jun 26

How Good Can Linear Models Be for Time-Series Forecasting?

arXiv:2606. 27282v1 Announce Type: new Abstract: Time-series forecasting research has been moving steadily toward larger architectures, from specialized transformers to general-purpose foundation models, on the assumption that capacity is what unlocks accuracy.

By Lang Huang, Jinglue Xu, Luke Darlow

llms benchmarks

Hugging Face Trending Papers

Aug 20

Systematic Evaluation of TabPFN-TS for Zero-Shot Probabilistic Heat Load Forecasting in District Heating Networks

District heating energy hubs require reliable heat load forecasts for efficient operational scheduling. Conventional forecasting workflows train system-specific models on historical data, which can become burdensome when networks change through new consumers, retrofits, or changing operating regimes.

arXiv Machine Learning

Sep 15

FlowTSFM: Turning Encoder Depth into Quantile Transport

arXiv:2609.13640v1 Announce Type: new Abstract: Encoder-based time series foundation models (TSFMs) typically rely on deep stacks of independently parameterized Transformer layers, where only the fin...

By Bahaeddine Abdessalem, Shifeng Xie, Zehao Xiao, Youssef Attia El Hili, Ambroise Odonnat, Jianfeng Zhang, Lujia Pan, Keli Zhang, Malik Tiomoko

llms

One email a morning, machine-written

Related stories

Adaptive Patching Is Harder Than It Looks For Time-Series Forecasting

Systematic Evaluation of TabPFN-TS and Chronos-2 for Zero-Shot Heat Load Forecasting in District Heating Networks

How Good Can Linear Models Be for Time-Series Forecasting?

How Good Can Linear Models Be for Time-Series Forecasting?

Systematic Evaluation of TabPFN-TS for Zero-Shot Probabilistic Heat Load Forecasting in District Heating Networks

FlowTSFM: Turning Encoder Depth into Quantile Transport