L-Drive: Beyond a Single Mapping-Latent Context Drives Time Series Forecasting

TL;DR

L-Drive is a change-aware time series forecasting framework that explicitly models high-level dynamics and adapts to regime shifts, improving accuracy and efficiency over traditional methods.

Contribution

It introduces a Latent-Context and gating mechanisms to enhance change detection and adaptation in multivariate time-series forecasting.

Findings

L-Drive outperforms existing methods in accuracy.

It achieves a better trade-off between accuracy and efficiency.

The framework effectively adapts to distribution shifts and regime changes.

Abstract

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.