Self-training Improves Pre-training for Natural Language Understanding

Abstract

Unsupervised pre-training has led to much recent progress in natural language understanding. In this paper, we study self-training as another way to leverage unlabeled data through semi-supervised learning. To obtain additional data for a specific task, we introduce SentAugment, a data augmentation method which computes task-specific query embeddings from labeled data to retrieve sentences from a bank of billions of unlabeled sentences crawled from the web. Unlike previous semi-supervised methods, our approach does not require in-domain unlabeled data and is therefore more generally applicable. Experiments show that self-training is complementary to strong RoBERTa baselines on a variety of tasks. Our augmentation approach leads to scalable and effective self-training with improvements of up to 2.6% on standard text classification benchmarks. Finally, we also show strong gains on knowledge-distillation and few-shot learning.

8 pages

References in corpus (6)

• Distilling the Knowledge in a Neural Network
• Language Models are Few-Shot Learners
• Billion-scale semi-supervised learning for image classification
• CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Data
• fairseq: A Fast, Extensible Toolkit for Sequence Modeling
• Simple and Effective Paraphrastic Similarity from Parallel Translations