Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Conversational Discourse Parsing on STAC (test)
Loading...
74.1
Arc Score
DiscProReco(w/o DPR)
68.38
69.865
71.35
72.835
Jun 7, 2021
Arc Score
Relation Score
Link F1
Link+Rel F1
Updated 4d ago
Evaluation Results
Method
Method
Links
Arc Score
Relation Score
Link F1
Link+Rel F1
DiscProReco(w/o DPR)
with DPR=false
2021.06
74.1
57
-
-
Deep Sequential
2021.06
73.2
55.7
-
-
Deep+MST
Architecture=GRU + MST
2021.06
69.6
52.1
-
-
Deep+Greedy
Decoding=Greedy
2021.06
69.3
51.9
-
-
Deep+ILP
Architecture=GRU + ILP
2021.06
69
53.1
-
-
MST
2021.06
68.8
50.4
-
-
ILP
2021.06
68.6
52.1
-
-
Deep Sequential Parser
Training data=Joint (S...
2021.10
-
-
72.8
54.8
Our Proposed Parser w/ language backbone
Training data=Joint (S...
2021.10
-
-
75.5
57.2
Feedback
Search any
task
Search any
task