mlsanityData to 14 Oct 2026

Downside-Controlled Online Forecast Combination under Delayed and Revised Outcomes

arXiv 2609.29096 ↗First posted 24 Sep 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Post-hoc correction adjusts a forecaster that cannot be retrained, such as a foundation model, but a correction fitted where errors are stable can hurt where they shift. We aim for downside control: not much worse than the starting forecast. We combine the frozen forecaster, a static corrector and an online corrector on the simplex, using only losses that mature after the horizon. Across seven benchmarks and four base models, two of them foundation models, the worst deterioration over 28 pairs at the main horizon is 0.15% and gains reach 11.5%. On day-ahead load for seven European bidding zones it lowers mean MSE in all seven zones, while single correctors raise mean MSE by up to 102% where the published forecast is most accurate. Three empirical conditions on expert speed, stream length and outcome alignment, each fixed by a documented failure, delimit its scope. Learning from the provisional outcome improves four zones on the settled one; learning on the settled outcome restores all seven.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 33 of its 78 references that are in the map. First posted on arXiv 24 Sep 2026.

Builds on

33 of 78 references.

References · 33

  1. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  2. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,169
  3. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,676
  4. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,424
  5. 2024Chronos: Learning the Language of Time Seriescited by 759
  6. 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 700
  7. 2023A decoder-only foundation model for time-series forecastingcited by 692
  8. 2023One Fits All:Power General Time Series Analysis by Pretrained LMcited by 632
  9. 2021Adaptive Conformal Inference Under Distribution Shiftcited by 375
  10. 2024Foundation Models for Time Series Analysis: A Tutorial and Surveycited by 239
  11. 2024Are Language Models Actually Useful for Time Series Forecasting?cited by 168
  12. 2017Optimizing the Latent Space of Generative Networkscited by 157
21 more
  1. 2013Online Learning under Delayed Feedbackcited by 141
  2. 2024CALF: Aligning LLMs for Time Series Forecasting via Cross-modal Fine-Tuningcited by 108
  3. 2023Conformal PID Control for Time Series Predictioncited by 88
  4. 2023OneNet: Enhancing Time Series Forecasting Models under Concept Drift by Online Ensemblingcited by 74
  5. 2014Optimal learning with Bernstein Online Aggregationcited by 35
  6. 2025Battling the Non-stationarity in Time Series Forecasting via Test-time Adaptationcited by 35
  7. 2022Forecast combinations: an over 50-year reviewcited by 34
  8. 2023Calibration of Time-Series Forecasting: Detecting and Adapting Context-Driven Distribution Shiftcited by 25
  9. 2025Improving Time Series Forecasting via Instance-aware Post-hoc Revisioncited by 15
  10. 2025Accurate Parameter-Efficient Test-Time Adaptation for Time Series Forecastingcited by 14
  11. 2024Minusformer: Improving Time Series Forecasting by Progressively Learning Residualscited by 10
  12. 2026The Forecast After the Forecast: A Post-Processing Shift in Time Seriescited by 10
  13. 2026Learning the Context of Errors: Black-Box Online Adaptation of Time Series Foundation Modelscited by 9
  14. 2022Residual Correction in Real-Time Traffic Forecastingcited by 8
  15. 2025Revisiting LLMs as Zero-Shot Time-Series Forecasters: Small Noise Can Break Large Modelscited by 7
  16. 2024Act Now: A Novel Online Forecasting Framework for Large-Scale Streaming Datacited by 6
  17. 2023On Forecast Stabilitycited by 5
  18. 2023Multivariate Probabilistic CRPS Learning with an Application to Day-Ahead Electricity Pricescited by 4
  19. 2024Using dynamic loss weighting to boost improvements in forecast stabilitycited by 4
  20. 2022A loss discounting framework for model averaging and selection in time series modelscited by 2
  21. 2025Energy load forecasting using Terna public data: a free lunch multi-task combination approachcited by 1

Built on by

Not cited yet.