Statements (63)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:benchmark
|
| gptkbp:aims_to |
language model performance
|
| gptkbp:based_on |
gptkb:GLUE
|
| gptkbp:challenges |
current models
|
| gptkbp:competes_with |
gptkb:GLUE
|
| gptkbp:consists_of |
multiple tasks
|
| gptkbp:developed_by |
gptkb:Allen_Institute_for_AI
|
| gptkbp:encourages |
research in NLP
|
| gptkbp:has_impact_on |
AI development
|
| https://www.w3.org/2000/01/rdf-schema#label |
Super GLUE
|
| gptkbp:is_adopted_by |
gptkb:film_production_company
|
| gptkbp:is_available_on |
gptkb:archive
|
| gptkbp:is_designed_for |
natural language understanding
|
| gptkbp:is_documented_in |
research papers
|
| gptkbp:is_evaluated_by |
F1 score
accuracy human annotators test sets |
| gptkbp:is_influenced_by |
previous benchmarks
|
| gptkbp:is_part_of |
NLP benchmarks
|
| gptkbp:is_related_to |
transfer learning
|
| gptkbp:is_supported_by |
community contributions
|
| gptkbp:is_tasked_with |
gptkb:Re_Co_RD
gptkb:CB gptkb:COPA RTE Bool Q Multi RC Wi C |
| gptkbp:is_used_by |
gptkb:physicist
|
| gptkbp:performance |
gptkb:Model
language translation dialog systems question answering reading comprehension sentiment analysis information retrieval pre-trained models transfer learning contextual embeddings multi-task learning text generation language modeling text summarization text classification named entity recognition part-of-speech tagging few-shot learning semantic understanding zero-shot learning coreference resolution text similarity text entailment language understanding tasks fine-tuning techniques commonsense reasoning |
| gptkbp:promotes |
robustness in models
|
| gptkbp:provides |
leaderboard
|
| gptkbp:released_in |
gptkb:2019
|
| gptkbp:requires |
contextual understanding
|
| gptkbp:bfsParent |
gptkb:GLUE_benchmark
gptkb:T5-3_B |
| gptkbp:bfsLayer |
4
|