Statements (53)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
Transformer model
|
| gptkbp:application |
Natural Language Processing
|
| gptkbp:architecturalStyle |
Encoder-Decoder
|
| gptkbp:availableIn |
Model_Hub
|
| gptkbp:communitySupport |
gptkb:PyTorch
TensorFlow Hugging Face Transformers |
| gptkbp:completed |
State-of-the-art results
|
| gptkbp:contributedTo |
Advancements in AI
OpenAI's research Google's_AI_initiatives Research_in_NLP |
| gptkbp:developedBy |
gptkb:Google_Research
|
| gptkbp:enhances |
Question answering
Sentiment analysis Text classification Translation Text summarization Named entity recognition Text generation Dialogue systems Text completion Text entailment |
| gptkbp:evaluates |
gptkb:BLEU
Accuracy F1 Score ROUGE |
| gptkbp:has |
Research papers
Tutorials Documentation Community forums Fine-tuning scripts Pre-trained weights |
| https://www.w3.org/2000/01/rdf-schema#label |
T5 Family
|
| gptkbp:improves |
Task-specific performance
|
| gptkbp:influencedBy |
GPT-2
BERT Transformer architecture XLNet |
| gptkbp:introduced |
2019
|
| gptkbp:model |
gptkb:T5-Base
gptkb:T5-3B gptkb:T5-Large gptkb:T5-11B T5-Small |
| gptkbp:powerOutput |
Text
|
| gptkbp:providesTrainingFor |
C4 dataset
|
| gptkbp:requires |
Large computational resources
|
| gptkbp:supports |
Multiple languages
Multi-task learning |
| gptkbp:trainingPrograms |
Causal Language Modeling
|
| gptkbp:uses |
Transfer Learning
Text-to-Text Transfer Transformer |