Constituency Parsing on Chinese Treebank 5.1 (test)
92.27F1 ScoreNeural CRF Parser
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Neural CRF ParserEmbeddings=BERT2020.08 | 92.27 | 92.04 | 92.51 | |
| Multilingual Self-Attentive Parsertraining_mode=monolingual2018.12 | 91.75 | 91.55 | 91.96 | |
| Kitaev et al. [2019]Embeddings=BERT2020.08 | 91.75 | 91.55 | 91.96 | |
| Neural CRF ParserPre-trained embeddings=true2020.08 | 89.8 | 89.89 | 89.71 | |
| Zhou and Zhao [2019]Pre-trained embeddings=true, Type=Structured skip-gram2020.08 | 89.4 | 89.09 | 89.7 | |
| Neural CRF ParserEmbeddings=Random2020.08 | 89.1 | 89.03 | 89.18 | |
| Kitaev and Klein [2018]Embeddings=Random2020.08 | 87.43 | 86.78 | 88.09 | |
| Teng and Zhang2018.12 | 87.3 | 87.1 | 87.5 | |
| Teng and Zhang [2018]2020.08 | 87.3 | 87.1 | 87.5 | |
| Fried and Klein2018.12 | 87 | — | — | |
| Shen et al. [2018]2020.08 | 86.5 | 86.4 | 86.6 | |
| Vilares et al. [2019]2020.08 | 85.61 | — | — | |
| Gomez-Rodriguez and Vilares [2018]2020.08 | 84.4 | — | — |