A Primer on Contrastive Pretraining in Language Processing: Methods, Lessons Learned and Perspectives

TL;DR

This paper reviews contrastive pretraining methods in NLP, discussing their applications, challenges, and lessons learned, aiming to bridge the gap with successful image representation techniques.

Contribution

It provides a comprehensive survey of contrastive NLP pretraining methods, highlighting key concepts, applications, and future research directions.

Findings

Contrastive NLP methods improve language modeling and task performance.

Automated text augmentation remains a major challenge in NLP contrastive learning.

Lessons from image pretraining can inform future NLP contrastive approaches.

Abstract

Modern natural language processing (NLP) methods employ self-supervised pretraining objectives such as masked language modeling to boost the performance of various application tasks. These pretraining methods are frequently extended with recurrence, adversarial or linguistic property masking, and more recently with contrastive learning objectives. Contrastive self-supervised training objectives enabled recent successes in image representation pretraining by learning to contrast input-input pairs of augmented images as either similar or dissimilar. However, in NLP, automated creation of text input augmentations is still very challenging because a single token can invert the meaning of a sentence. For this reason, some contrastive NLP pretraining methods contrast over input-label pairs, rather than over input-input pairs, using methods from Metric Learning and Energy Based Models. In this survey, we summarize recent self-supervised and supervised contrastive NLP pretraining methods and describe where they are used to improve language modeling, few or zero-shot learning, pretraining data-efficiency and specific NLP end-tasks. We introduce key contrastive learning concepts with lessons learned from prior research and structure works by applications and cross-field relations. Finally, we point to open challenges and future directions for contrastive NLP to encourage bringing contrastive NLP pretraining closer to recent successes in image representation pretraining.