Open AccessJournal Article
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Colin Raffel,Noam Shazeer,Adam Roberts,Katherine Lee,Sharan Narang,Michael Matena,Yanqi Zhou,Wei Li,Peter J. Liu +8 more
TLDR
This article introduced a unified framework that converts all text-based language problems into a text-to-text format and compared pre-training objectives, architectures, unlabeled data sets, transfer approaches, and other factors on dozens of language understanding tasks.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.read more
Citations
More filters
Proceedings ArticleDOI
COM2SENSE: A Commonsense Reasoning Benchmark with Complementary Sentences
TL;DR: In this article, the authors introduce a commonsense reasoning benchmark dataset comprising natural language true/false statements, with each sample paired with its complementary counterpart, resulting in 4k sentence pairs.
Proceedings Article
proScript: Partially Ordered Scripts Generation
TL;DR: This paper used pre-trained neural language models to generate high-quality scripts, at varying levels of granularity, for a wide range of everyday scenarios (e.g., bake a cake).
Book ChapterDOI
Generating Empathetic Responses with a Pre-trained Conversational Model
TL;DR: In this article, a pre-trained neural conversational language model named DialoGPT and a new collection of empathetic dialogues tagged with emotions are used in order to investigate the ability of the model in learning and generating more empathic responses.
Proceedings ArticleDOI
PASS: Perturb-and-Select Summarizer for Product Reviews
Nadav Oved,Ran Levy +1 more
TL;DR: Perturb-and-Select Summarizer (P PASS) as mentioned in this paper employs a large pre-trained Transformer-based model, which follows a few-shot fine-tuning scheme.
Posted Content
Retrieval-guided Counterfactual Generation for QA
TL;DR: This paper developed a Retrieve-Generate-Filter (RGF) technique to create counterfactual evaluation and training data with minimal human supervision, using an open-domain QA framework and question generation model trained on original task data.