Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
At a glance
T5 treats language tasks as text input and text output within one framework.
In one paragraph
Introduced T5, a model that treats language tasks as text in and text out, from translation to answering questions. A systematic comparison of training methods, alongside the new C4 text dataset, helped establish a reusable approach to training models for many language tasks.
Authors 9
- Colin Raffel
- Noam Shazeer
- Adam Roberts
- Katherine Lee
- Sharan Narang
- Michael Matena
- Yanqi Zhou
- Wei Li
- Peter J. Liu
