T5 Research Publication
Google researchers submitted the T5 paper, 'Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer,' on October 23, 2019, introducing a unified text-to-text framework for transfer learning in NLP.
Historical classification
This verified record belongs to LINEAiGE's artificial-intelligence history layer beginning in 1956.
Why it matters
A documented language-model research milestone that framed diverse text-based NLP tasks within a single text-to-text formulation and released supporting data, pretrained models, and code for reproducibility.
People, organizations & technologies
PEOPLE
Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, Peter J. Liu
ORGANIZATIONS
Google
TECHNOLOGIES
T5, Transformer, C4
Evidence & verification
STATE
verified
CONFIDENCE
high
LAST VERIFIED
2026-09-10T01:03:41+08:00
The primary paper was submitted to arXiv on October 23, 2019 and explicitly introduces a unified framework that converts text-based language problems into a text-to-text format, together with the Colossal Clean Crawled Corpus and released pretrained models/code. Google Research later described the resulting model as the Text-To-Text Transfer Transformer (T5) and reported an 11-billion-parameter largest model. This record intentionally uses the dated 2019 paper submission as the event boundary. It does not reinterpret that date as a commercial product launch, API release, deployment date, unrestricted availability milestone, geography claim, or evidence of causal influence on later systems. No ancestry, influence, causality, geography, or relationships are inferred.
Sources
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
arXiv
PUBLISHED
2019-10-23
Exploring Transfer Learning with T5: the Text-To-Text Transfer Transformer
Google Research
PUBLISHED
2020-02-24
Trace the chronology
Record identity
LINEAiGE IDt5-2019