SOTA Constituency Parsing on Penn Treebank WSJ (section 23 test) and PapersWithCode

95.8F1 Score

Berkeley Neural

Updated 3d ago

Evaluation Results

Method	Links
Berkeley Neural 2025.12		95.8
MetaJuLS 2025.12		95.6
A* Neural 2025.12		95.2
Supervised Ranking 2025.12		94.6
Activity-Based 2025.12		94.5
VSIDS-style 2025.12		94.3
In-order parser 2017.07		94.2
Cost-Normalized Greedy 2025.12		94.2
GPU CKY 2025.12		94.1
Choe and Charniak 2017.07		93.8
Kuncoro et al. 2017.07		93.6
In-order parser 2017.07		93.6
Dyer et al. 2017.07		93.3
Top-down parser 2017.07		93.3
Bottom-up parser 2017.07		93.3
LSTM+A ensemble 2014.12		92.8
Choe and Charniak 2017.07		92.6
LSTM+A 2014.12		92.5
Huang & Harper (2010) ensemble 2014.12		92.4
Shindo et al. (2012) 2016.02		92.4
RNNG 2016.02		92.4
McClosky et al. (2006) 2014.12		92.1
McClosky et al. (2006) 2016.02		92.1
Vinyals et al. (2015) 2016.02		92.1
Petrov et al. (2010) ensemble 2014.12		91.8
In-order parser 2017.07		91.8
Liu and Zhang 2017.07		91.7
Huang 2017.07		91.7
Charniak and Johnson 2017.07		91.5
Zhu et al. (2013) 2014.12		91.3
Huang & Harper (2009) 2014.12		91.3
Zhu et al. (2013) 2016.02		91.3
Cross and Huang 2017.07		91.3
Bottom-up parser 2017.07		91.3
Dyer et al. 2017.07		91.2
Top-down parser 2017.07		91.2
Shindo et al. (2012) 2016.02		91.1
Shindo et al. 2017.07		91.1
Durrett and Klein 2017.07		91.1
Bod (2003) 2016.02		90.7
Vinyals et al. 2017.07		90.7
Watanabe and Sumita 2017.07		90.7
LSTM+A+D ensemble 2014.12		90.5
baseline LSTM 2014.12		90.5
Petrov et al. (2006) 2014.12		90.4
Zhu et al. (2013) 2014.12		90.4
Socher et al. (2013a) 2016.02		90.4
Zhu et al. (2013) 2016.02		90.4
Socher et al. 2017.07		90.4
Zhu et al. 2017.07		90.4
Petrov and Klein (2007) 2016.02		90.1
RNNG 2016.02		89.8
Henderson (2004) 2016.02		89.4
LSTM+A+D 2014.12		88.3
Vinyals et al. (2015) 2016.02		88.3