Statements (51)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
benchmark
|
| gptkbp:aimsTo |
generalization ability
|
| gptkbp:composedOf |
nine tasks
|
| gptkbp:designedFor |
natural language understanding
|
| gptkbp:hasImpactOn |
NLP_research
|
| https://www.w3.org/2000/01/rdf-schema#label |
GLUE benchmark
|
| gptkbp:includes |
gptkb:SST-2
CoLA MNLI QNLI WNLI RTE QQP STS-B |
| gptkbp:introduced |
2018
|
| gptkbp:isAvailableIn |
gluebenchmark.com
|
| gptkbp:isChallengedBy |
gptkb:DistilBERT
gptkb:ERNIE gptkb:T5 GPT-2 BERT ALBERT transformer models SuperGLUE RoBERTa XLNet DeBERTa ELECTRA |
| gptkbp:isCitedIn |
numerous research papers
|
| gptkbp:isEvaluatedBy |
gptkb:Spearman's_rank_correlation
F1 score accuracy Matthew's_correlation_coefficient |
| gptkbp:isPartOf |
NLP_evaluation_frameworks
|
| gptkbp:isPopularIn |
machine learning community
|
| gptkbp:isRelatedTo |
pre-trained models
transfer learning fine-tuning language models evaluation benchmarks |
| gptkbp:isSupportedBy |
gptkb:PyTorch
gptkb:Keras TensorFlow Hugging Face Transformers |
| gptkbp:isUsedFor |
hyperparameter tuning
model selection performance comparison |
| gptkbp:provides |
evaluation metrics
|
| gptkbp:relatedModel |
various_NLP_tasks
|
| gptkbp:releasedIn |
AI2
|
| gptkbp:usedBy |
researchers
|