Statements (134)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:language
|
| gptkbp:application |
natural language processing
|
| gptkbp:applies_to |
natural language processing
|
| gptkbp:architecture |
gptkb:Transformers
|
| gptkbp:based_on |
gptkb:Transformers
|
| gptkbp:can_be_used_in |
question answering
sentiment analysis text generation text classification named entity recognition |
| gptkbp:developed_by |
gptkb:Google_Brain
|
| gptkbp:evaluates |
gptkb:GLUE_benchmark
|
| gptkbp:field_of_study |
gptkb:Artificial_Intelligence
|
| gptkbp:has |
multiple pre-trained models
|
| gptkbp:has_achieved |
state-of-the-art results
|
| https://www.w3.org/2000/01/rdf-schema#label |
XLNet
|
| gptkbp:improves |
gptkb:BERT
|
| gptkbp:input_output |
predicted tokens
sequence of tokens |
| gptkbp:introduced_in |
gptkb:2019
|
| gptkbp:is_adopted_by |
government agencies
academic institutions startups tech companies research labs |
| gptkbp:is_available_on |
gptkb:Git_Hub
|
| gptkbp:is_compared_to |
gptkb:GPT-2
gptkb:T5 gptkb:Ro_BERTa ALBERT ERNIE |
| gptkbp:is_designed_for |
language understanding tasks
|
| gptkbp:is_documented_in |
research papers
tutorials online courses technical blogs |
| gptkbp:is_evaluated_by |
gptkb:MNLI_benchmark
gptkb:SQu_AD_benchmark gptkb:GLUE_benchmark gptkb:Co_LA_benchmark gptkb:QQP_benchmark gptkb:RACE_benchmark gptkb:WNLI_benchmark STS-B benchmark |
| gptkbp:is_open_source |
gptkb:true
|
| gptkbp:is_part_of |
NLP research community
|
| gptkbp:is_supported_by |
gptkb:Tensor_Flow
gptkb:Py_Torch |
| gptkbp:is_trained_in |
large text corpora
|
| gptkbp:is_used_in |
language translation
dialog systems search engines chatbots virtual assistants content generation text summarization |
| gptkbp:language |
English
|
| gptkbp:orbital_period |
340 million
|
| gptkbp:outperforms |
traditional autoregressive models
|
| gptkbp:performance |
state-of-the-art
|
| gptkbp:pretraining_objective |
permutation language modeling
|
| gptkbp:provides_information_on |
large text corpus
|
| gptkbp:related_to |
gptkb:GPT-2
gptkb:T5 gptkb:Ro_BERTa |
| gptkbp:release_year |
gptkb:2019
|
| gptkbp:successor |
gptkb:BERT
|
| gptkbp:supports |
unsupervised learning
|
| gptkbp:tokenization_method |
Word Piece
|
| gptkbp:training |
unsupervised learning
|
| gptkbp:tuning |
transfer learning
|
| gptkbp:use_case |
gptkb:market_research
gptkb:academic_research gptkb:Biology gptkb:Writer language learning language translation business intelligence data visualization data-driven decision making dialog systems question answering content creation contextual advertising data science sentiment analysis text analysis topic modeling chatbots data augmentation educational tools information retrieval semantic search data storytelling social media monitoring text generation text mining language modeling voice assistants code generation text summarization content recommendation e-learning text classification text completion brand monitoring language understanding named entity recognition customer support automation personal assistants style transfer semantic role labeling paraphrasing knowledge extraction text entailment fake news detection multi-turn dialogue user intent detection |
| gptkbp:uses |
permutation language modeling
|
| gptkbp:utilizes |
self-attention mechanism
|
| gptkbp:bfsParent |
gptkb:MNLI_benchmark
gptkb:T5-large gptkb:Attention_Is_All_You_Need gptkb:GPT-2 gptkb:GPT-3 gptkb:GLUE_benchmark gptkb:T5 gptkb:Transformer_Models gptkb:SQu_AD gptkb:Hugging_Face gptkb:Super_GLUE_benchmark gptkb:T5-3_B gptkb:Transformer_Architecture |
| gptkbp:bfsLayer |
5
|