← Back to all news
arXiv AI By Yuchen He, Yueyang Cang, Zhiyuan Ning, Ningyu Wang, Li Shi

RATL: Learning from Retrieved Residuals for Robust Multivariate Time-Series Forecasting

Read the original on arXiv AI →

RATL is a plug‑in method for multivariate time‑series forecasting that uses a frozen base forecaster to build a memory of its historical forecast residuals. During inference, RATL retrieves residual trajectories from similar past contexts and employs a set‑aware router to combine them, providing learned feedback correction. Experiments demonstrate that this residual‑retrieval approach improves the performance of the base forecaster across various benchmarks and backbones.

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 AI.

Related stories

arXiv Machine Learning
Jun 10

One Step Closer to Ground Truth: A Multi-Scale Residual-Aware Representation Learning Pipeline for Predicting Time Series Data

arXiv:2606. 10678v1 Announce Type: new Abstract: Transformer-based models have emerged as leading paradigms in time-series forecasting in recent years, employing self-attention mechanisms to capture long-range dependencies.

By Amrijit Biswas, Mustafa Kamal, Robin Krambroeckers, M. M. Lutfe Elahi, Sifat Momen, Nabeel Mohammed, Shafin Rahman
llmsbenchmarkssafety
More like this →
arXiv AI
Oct 19

Look Back, Correct Forward: Error Feedback Calibration for Time Series Forecasting

The paper introduces Error Feedback Calibration (EFC), a test‑time framework that corrects deterministic time‑series forecasts by using revealed errors as causal feedback. EFC builds a Causal Error Diagonal (CED) from overlapping historical forecast errors, applies Cumulative Error Transport (CET) to map this error state to future residuals, and employs Temporal Residual Learning (TRL) to predict remaining residual structure. Experiments across various forecasting backbones and datasets demonstrate that EFC improves performance while limiting over‑correction.

By Haonan Yang, Yulin He, Zhuo Li
More like this →
arXiv AI
Jun 16

TS-Memory: Plug-and-Play Memory for Time Series Foundation Models

arXiv:2602. 11550v2 Announce Type: replace-cross Abstract: Time Series Foundation Models (TSFMs) achieve strong zero-shot forecasting through large-scale pre-training, but adapting them to downstream domains under distribution shift remains challenging.

By Sisuo Lyu, Siru Zhong, Tiegang Chen, Weilin Ruan, Qingxiang Liu, Taiqiang Lv, Qingsong Wen, Raymond Chi-Wing Wong, Yuxuan Liang
efficiencybenchmarks
More like this →
arXiv AI
Aug 18

AsyTO: Asymmetric Temporal Operator for Parameter-Efficient Multivariate Time Series Forecasting

arXiv:2608. 16098v1 Announce Type: cross Abstract: Multivariate time-series forecasting faces a structural dilemma: sharing one temporal predictor across variables is parameter-efficient but forces heterogeneous variables through an identical history-to-future map, whereas learning an independent predictor per variable restores flexibility at a cost that grows with the product of variable count, context length, and horizon.

By Xiachong Lin, Du Yin, Hao Xue, Wen Hu, Imran Razzak, Arian Prabowo, Matthew Amos, Flora D. Salim
benchmarks
More like this →