Bilingual Rhetorical Structure Parsing with Large Parallel Annotations
Open paper
Benchmark
Model
Rank
Results
discourse-parsing-on-rst-dt
DMRST
#5
Standard Parseval (Full): 55.7 ± 0.3
Standard Parseval (Span): 78.7 ± 0.4
…
end-to-end-rst-parsing-on-rst-dt
DMRST + ToNy + E-BiLSTM
#1
Standard Parseval (Full): 53.0 ± 0.7
…
Rank counts only results with a code link.