This is not the latest version of this item. The latest version can be found here.
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
No Thumbnail Available
Date
2019-10-19
Journal Title
Journal ISSN
Volume Title
Publisher
arXiv
Abstract
Transfer learning, where a model is first pre-trained on a data-rich task before being fine-tuned on a downstream task, has emerged as a powerful technique in natural language processing (NLP). The effectiveness of transfer learning has given rise to a diversity of approaches, methodology, and practice. In this paper, we explore the landscape of transfer learning techniques for NLP by introducing a unified framework that converts all text-based language problems into a text-to-text format. Our systematic study compares pre-training objectives, architectures, unlabeled data sets, transfer approaches, and other factors on dozens of language understanding tasks. By combining the insights from our exploration with scale and our new ``Colossal Clean Crawled Corpus'', we achieve state-of-the-art results on many benchmarks covering summarization, question answering, text classification, and more. To facilitate future work on transfer learning for NLP, we release our data set, pre-trained models, and code.
Description
이 논문은 모든 NLP 태스크를 text-to-text 문제로 재구성하는 T5 프레임워크를 제안하며, 전이학습의 효과를 대규모 데이터셋에서 실험적으로 검증합니다.
©2019 Google Research