Multi-Task Learning in Natural Language Processing: An Overview

TL;DR

This paper provides a comprehensive overview of Multi-Task Learning (MTL) in NLP, covering architectures, optimization techniques, applications, benchmarks, and future research directions.

Contribution

It categorizes MTL architectures in NLP, reviews optimization methods, and discusses diverse applications and datasets, offering a thorough survey of the field.

Findings

MTL architectures include parallel, hierarchical, modular, and generative adversarial types.

Optimization techniques like loss construction and task scheduling improve multi-task training.

MTL enhances performance across various NLP tasks.

Abstract

Deep learning approaches have achieved great success in the field of Natural Language Processing (NLP). However, directly training deep neural models often suffer from overfitting and data scarcity problems that are pervasive in NLP tasks. In recent years, Multi-Task Learning (MTL), which can leverage useful information of related tasks to achieve simultaneous performance improvement on these tasks, has been used to handle these problems. In this paper, we give an overview of the use of MTL in NLP tasks. We first review MTL architectures used in NLP tasks and categorize them into four classes, including parallel architecture, hierarchical architecture, modular architecture, and generative adversarial architecture. Then we present optimization techniques on loss construction, gradient regularization, data sampling, and task scheduling to properly train a multi-task model. After presenting applications of MTL in a variety of NLP tasks, we introduce some benchmark datasets. Finally, we make a conclusion and discuss several possible research directions in this field.