Constituency Parsing on PTB Diagnostic ECG Database

Metric: Max F1 (WSJ) (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	Max F1 (WSJ)▼	Extra Data	Paper	Date↕	Code
1	Ensemble (Generative MBR)	71.9	No	Ensemble Distillation for Unsupervised Constitue...	2023-10-03	Code
2	Parse-Focused (NT=4500)	70.3	No	Structural Optimization Ambiguity and Simplicity...	2024-07-23	Code
3	Parse-Focused (NT=30)	68.4	No	Structural Optimization Ambiguity and Simplicity...	2024-07-23	Code
4	inside-outside co-training + weak supervision	66.8	Yes	Co-training an Unsupervised Constituency Parser ...	2021-10-05	Code
5	Hashing (Parserker 2)	64.1	No	On Eliciting Syntax from Language Models via Has...	2024-10-05	Code
6	S-DIORA	63.96	No	-	-	-
7	TN-PCFG (p=500)	61.4	No	PCFGs Can Do Better: Inducing Probabilistic Cont...	2021-04-28	Code
8	Compound PCFG	60.1	No	Compound Probabilistic Context-Free Grammars for...	2019-06-24	Code
9	DIORA (+PP)	56.2	No	-	-	Code
10	Neural PCFG	52.6	No	Compound Probabilistic Context-Free Grammars for...	2019-06-24	Code
11	URNNG	52.4	No	Unsupervised Recurrent Neural Network Grammars	2019-04-07	Code
12	ON-LSTM (tuned)	50	No	Ordered Neurons: Integrating Tree Structures int...	2018-10-22	Code
13	DIORA	49.6	No	-	-	Code
14	ON-LSTM	49.4	No	Ordered Neurons: Integrating Tree Structures int...	2018-10-22	Code
15	PRPN (tuned)	47.9	No	Neural Language Modeling by Jointly Learning Syn...	2017-11-02	Code
16	PRPN	38.1	No	Neural Language Modeling by Jointly Learning Syn...	2017-11-02	Code

#1Ensemble (Generative MBR)SOTA
71.9
Max F1 (WSJ)· 2023-10-03
Ensemble Distillation for Unsupervised Constituency Parsing Code
#2Parse-Focused (NT=4500)
70.3
Max F1 (WSJ)· 2024-07-23
Structural Optimization Ambiguity and Simplicity Bias in Unsupervised Neural Grammar Induction Code
#3Parse-Focused (NT=30)
68.4
Max F1 (WSJ)· 2024-07-23
Structural Optimization Ambiguity and Simplicity Bias in Unsupervised Neural Grammar Induction Code
#4inside-outside co-training + weak supervisionSOTA
66.8
Max F1 (WSJ)· Extra Data· 2021-10-05
Co-training an Unsupervised Constituency Parser with Weak Supervision Code
#5Hashing (Parserker 2)
64.1
Max F1 (WSJ)· 2024-10-05
On Eliciting Syntax from Language Models via Hashing Code
#6S-DIORA
63.96
Max F1 (WSJ)
No paper
#7TN-PCFG (p=500)SOTA
61.4
Max F1 (WSJ)· 2021-04-28
PCFGs Can Do Better: Inducing Probabilistic Context-Free Grammars with Many Symbols Code
#8Compound PCFGSOTA
60.1
Max F1 (WSJ)· 2019-06-24
Compound Probabilistic Context-Free Grammars for Grammar Induction Code
#9DIORA (+PP)
56.2
Max F1 (WSJ)
No paperCode
#10Neural PCFG
52.6
Max F1 (WSJ)· 2019-06-24
Compound Probabilistic Context-Free Grammars for Grammar Induction Code
#11URNNGSOTA
52.4
Max F1 (WSJ)· 2019-04-07
Unsupervised Recurrent Neural Network Grammars Code
#12ON-LSTM (tuned)SOTA
50
Max F1 (WSJ)· 2018-10-22
Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks Code
#13DIORA
49.6
Max F1 (WSJ)
No paperCode
#14ON-LSTM
49.4
Max F1 (WSJ)· 2018-10-22
Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks Code
#15PRPN (tuned)SOTA
47.9
Max F1 (WSJ)· 2017-11-02
Neural Language Modeling by Jointly Learning Syntax and Lexicon Code
#16PRPN
38.1
Max F1 (WSJ)· 2017-11-02
Neural Language Modeling by Jointly Learning Syntax and Lexicon Code