Constituency Parsing On Penn Treebank

Metrics

F1 score

Results

Performance results of various models on this benchmark

Model Name	F1 score	Paper Title	Repository
Model combination	94.66	Improving Neural Parsing by Disentangling Model Combination and Reranking Effects	-
Label Attention Layer + HPSG + XLNet	96.38	Rethinking Self-Attention: Towards Interpretability in Neural Parsing
RNN Grammar	93.3	Recurrent Neural Network Grammars
Tetra Tagging	95.44	Tetra-Tagging: Word-Synchronous Parsing with Linear-Time Inference
Head-Driven Phrase Structure Grammar Parsing (Joint) + XLNet	96.33	Head-Driven Phrase Structure Grammar Parsing on Penn Treebank
Hashing + XLNet	96.43	To be Continuous, or to be Discrete, Those are Bits of Questions
NFC + BERT-large	95.92	Investigating Non-local Features for Neural Constituency Parsing
Stack-only RNNG	93.6	What Do Recurrent Neural Network Grammars Learn About Syntax?
CRF Parser + RoBERTa	96.32	Fast and Accurate Neural CRF Constituency Parsing
Transformer	92.7	Attention Is All You Need
Self-attentive encoder + ELMo	95.13	Constituency Parsing with a Self-Attentive Encoder
SAPar + XLNet	96.40	Improving Constituency Parsing with Span Attention
Semi-supervised LSTM-LM	93.8	-	-
CNN Large + fine-tune	95.6	Cloze-driven Pretraining of Self-attention Networks	-
N-ary semi-markov + BERT-large	95.92	N-ary Constituent Tree Parsing with Recursive Semi-Markov Model
Head-Driven Phrase Structure Grammar Parsing (Joint) + BERT	95.84	Head-Driven Phrase Structure Grammar Parsing on Penn Treebank
Parse fusion	92.6	-	-
LSTM Encoder-Decoder + LSTM-LM	94.47	Direct Output Connection for a High-Rank Language Model
Attach-Juxtapose Parser + XLNet	96.34	Strongly Incremental Constituency Parsing with Graph Neural Networks
Self-training	92.1	Effective Self-Training for Parsing

0 of 27 row(s) selected.