Deep Bidirectional Language-Knowledge Graph Pretraining
Open paper
Benchmark
Model
Rank
Results
riddle-sense-on-riddle-sense
DRAGON
#1
Accuracy (%): 71.3
common-sense-reasoning-on-commonsenseqa
DRAGON
#9
Accuracy: 78.2
question-answering-on-medqa-usmle
DRAGON + BioLinkBERT
#16
Accuracy: 47.5
Rank counts only results with a code link.