T5 — Text-to-Text Transfer Transformer

Fact

A Transformer model handling many language tasks as text in and text out.

In Plain Words

T5 is like a drive-thru window for words. You hand over text. It hands back text.

It uses one setup for many language jobs. It can summarize, translate, and answer questions.

Related Concepts

• Transformer architecture
• Sequence-to-Sequence Model
• Pretraining
• Fine-tuning

Transformer
T5 is an encoder-decoder Transformer model.

Seq2Seq
T5 turns many language jobs into Seq2Seq jobs.

Pretraining
T5 first learns language skills from large text pretraining.

Fine-tuning
Fine-tuning adapts T5 for summary, translation, and Q&A jobs.