K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapters
Open paper
Benchmark
Model
Rank
Results
relation-classification-on-tacred-1
RoBERTa
#8
F1: 71.3
relation-classification-on-tacred-1
K-Adapter
#13
F1: 72.0
relation-extraction-on-tacred
K-ADAPTER (F+L)
#10
F1: 72.04
F1 (10% Few-Shot): 56.0
F1 (5% Few-Shot): 45.1
…
Rank counts only results with a code link.