Pushing on Text Readability Assessment: A Transformer Meets Handcrafted Linguistic Features
Open paper
Benchmark
Model
Rank
Results
text-classification-on-onestopenglish
RoBERTa-RF-T1 hybrid
#1
Accuracy (5-fold): 0.990
text-classification-on-weebit-readability
BART-RF-T1 hybrid
#2
Accuracy (5-fold): 0.905
Rank counts only results with a code link.