Time Waits for No One! Analysis and Challenges of Temporal Misalignment

TL;DR

This paper investigates how temporal misalignment affects NLP model performance across various domains and tasks, revealing significant impacts and the limited effectiveness of current adaptation methods.

Contribution

It introduces a comprehensive suite of eight diverse tasks to quantify temporal misalignment effects and compares adaptation strategies, highlighting the need for improved temporal robustness in NLP.

Findings

Temporal misalignment significantly degrades performance.

Continued pretraining offers limited improvements.

Task-specific finetuning on target data is most effective.

Abstract

When an NLP model is trained on text data from one time period and tested or deployed on data from another, the resulting temporal misalignment can degrade end-task performance. In this work, we establish a suite of eight diverse tasks across different domains (social media, science papers, news, and reviews) and periods of time (spanning five years or more) to quantify the effects of temporal misalignment. Our study is focused on the ubiquitous setting where a pretrained model is optionally adapted through continued domain-specific pretraining, followed by task-specific finetuning. We establish a suite of tasks across multiple domains to study temporal misalignment in modern NLP systems. We find stronger effects of temporal misalignment on task performance than have been previously reported. We also find that, while temporal adaptation through continued pretraining can help, these gains are small compared to task-specific finetuning on data from the target time period. Our findings motivate continued research to improve temporal robustness of NLP models.