Share your thoughts, 1 month free Claude Pro on usSee more

Part-of-Speech Tagging on WSJ (test)

97.78Accuracy

Ling et al.

Updated 1mo ago

Evaluation Results

Method	Links
Ling et al. 2017.11		97.78	-
LM-LSTM-CRF 2017.09		97.59	-
Adversarial POS Tagging 2017.11		97.58	-
Ma and Hovy 2018.06		97.55	-
Yang et al. 2018.06		97.55	-
LM-LSTM-CRF 2017.09		97.55	-
Ma and Hovy 2017.11		97.55	-
Yang et al. 2017.11		97.55	-
Hashimoto et al. 2017.11		97.55	-
BiLSTM-CRF 2017.11		97.54	-
LM-LSTM-CRF 2017.09		97.53	-
Lample et al. 2018.06		97.51	-
Lample et al. 2016 2017.09		97.51	-
Søgaard 2011 2017.09		97.5	-
Søgaard 2017.11		97.5	-
Src 2018.04		97.5	-
CLSTM+WLSTM+CRF 2018.06		97.49	-
CCNN+WLSTM+CRF 2018.06		97.46	-
Ma et al. 2016 2017.09		97.46	-
Tri 2018.04		97.45	-
Yang et al. 2017 2017.09		97.43	-
Stanford* 2018.04		97.43	-
Ma et al. 2016 2017.09		97.42	-
CLSTM+WCNN+CRF 2018.06		97.38	-
MT-Tri 2018.04		97.37	-
Bi-LSTM 2016.04		97.36	-
Sun 2014 2017.09		97.36	-
Lample et al. 2016 2017.09		97.35	-
CCNN+WCNN+CRF 2018.06		97.33	-
Convnet 2016.04		97.32	-
Collobert et al. 2011 2017.09		97.29	-
Collobert et al. 2017.11		97.29	-
Manning 2011 2017.09		97.28	-
Manning 2017.11		97.28	-
Toutanova et al. 2017.11		97.27	-
bi-LSTM (w+c) 2016.04		97.22	-
Nochar+WLSTM+CRF 2018.06		97.2	-
Asym 2018.04		97.19	-
Rei 2017 2017.09		97.14	-
FLORS* 2018.04		97.11	-
Nochar+WCNN+CRF 2018.06		96.99	-
Rei 2017 2017.09		96.97	-
Context-to-Vec 2022.10		96.91	92.8
BERT+Skip-gram 2022.10		96.86	92.3
Convnet (reimplementation) 2016.04		96.8	-
BERTavg 2022.10		96.8	90.76
SynGCN 2022.10		96.71	92.23
TnT* 2018.04		96.57	-
GloVe 2022.10		96.52	91.84
Bi-RNN 2016.04		95.93	-
Skip-gram 2022.10		95.12	89.03