Saved in:
Bibliographic Details
Main Authors: Zhang, Fan, Chen, Shijun, Wang, Hua
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.17730
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914598748160000
author Zhang, Fan
Chen, Shijun
Wang, Hua
author_facet Zhang, Fan
Chen, Shijun
Wang, Hua
contents Mainstream methods for multivariate time-series forecasting largely follow the Direct-Mapping paradigm. They learn a unified mapping from history to the future in the observation space to fit value-level dependencies. However, real-world systems often undergo distribution shifts and regime changes. In such cases, a unified mapping can exhibit response lag around turning points, causing error accumulation within the switching window and reducing forecasting reliability. To address this issue, we propose L-Drive, a change-aware forecasting framework. L-Drive introduces a Latent-Context, to explicitly characterize high-level dynamics evolving over time, and uses gating to modulate increment representations. This provides more timely change cues and improves adaptation to changing segments. In addition, it incorporates patch-shared relative positional basis functions to strengthen intra-segment structural modeling and reduce overfitting caused by absolute-position memorization. Extensive experiments validate the effectiveness of L-Drive and show a better overall trade-off between forecasting accuracy and computational efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2605_17730
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle L-Drive: Beyond a Single Mapping-Latent Context Drives Time Series Forecasting
Zhang, Fan
Chen, Shijun
Wang, Hua
Machine Learning
Artificial Intelligence
Mainstream methods for multivariate time-series forecasting largely follow the Direct-Mapping paradigm. They learn a unified mapping from history to the future in the observation space to fit value-level dependencies. However, real-world systems often undergo distribution shifts and regime changes. In such cases, a unified mapping can exhibit response lag around turning points, causing error accumulation within the switching window and reducing forecasting reliability. To address this issue, we propose L-Drive, a change-aware forecasting framework. L-Drive introduces a Latent-Context, to explicitly characterize high-level dynamics evolving over time, and uses gating to modulate increment representations. This provides more timely change cues and improves adaptation to changing segments. In addition, it incorporates patch-shared relative positional basis functions to strengthen intra-segment structural modeling and reduce overfitting caused by absolute-position memorization. Extensive experiments validate the effectiveness of L-Drive and show a better overall trade-off between forecasting accuracy and computational efficiency.
title L-Drive: Beyond a Single Mapping-Latent Context Drives Time Series Forecasting
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2605.17730