Accurate Parameter-Efficient Test-Time Adaptation for Time Series Forecasting

Heitor R. Medeiros, Hossein Sharifi-Noghabi, Gabriel L. Oliveira, Saghar Irandoust
6/29/2025
cs.LGcs.AI

Official code: BorealisAI/PETSA (21 stars, last commit 2025-05-30).

Abstract

Real-world time series often exhibit a non-stationary nature, degrading the performance of pre-trained forecasting models. Test-Time Adaptation (TTA) addresses this by adjusting models during inference, but existing methods typically update the full model, increasing memory and compute costs. We propose PETSA, a parameter-efficient method that adapts forecasters at test time by only updating small calibration modules on the input and output. PETSA uses low-rank adapters and dynamic gating to adjust representations without retraining. To maintain accuracy despite limited adaptation capacity, we introduce a specialized loss combining three components: (1) a robust term, (2) a frequency-domain term to preserve periodicity, and (3) a patch-wise structural term for structural alignment. PETSA improves the adaptability of various forecasting backbones while requiring fewer parameters than baselines. Experimental results on benchmark datasets show that PETSA achieves competitive or better performance across all horizons. Our code is available at: https://github.com/BorealisAI/PETSA

View on arXivView PDF

Code Implementations(1)

BorealisAI/PETSAOfficial100%

[PUT ICML 2025] Accurate Parameter-Efficient Test-Time Adaptation for Time Series Forecasting

213Shell, PythonMay 29, 20251 years agoNOASSERTION
deeplearningtest-time-adaptationtime-seriestime-series-forecasting

Similar Papers

Test-Time Adaptation for Non-stationary Time Series: From Synthetic Regime Shifts to Financial Markets

87% match2026

Battling the Non-stationarity in Time Series Forecasting via Test-time Adaptation

87% match2 repos462025

Learning with Calibration: Exploring Test-Time Computing of Spatio-Temporal Forecasting

85% match1 repo202025

Test Time Learning for Time Series Forecasting

85% match2024

AdaNODEs: Test Time Adaptation for Time Series Forecasting Using Neural ODEs

85% match2026

Bag of Tricks for Fully Test-Time Adaptation

83% match1 repo102023

Cite this paper

@article{medeiros2025accurate,
  title  = {Accurate Parameter-Efficient Test-Time Adaptation for Time Series Forecasting},
  author = {Heitor R. Medeiros and Hossein Sharifi-Noghabi and Gabriel L. Oliveira and Saghar Irandoust},
  year   = {2025},
  eprint = {2506.23424},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2506.23424v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2506.23424.svg)](https://codeofpaper.com/paper/2506.23424)

Discussion