CRF Parser + BERT

Reported on 3 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing3 results

Constituency ParsingonCTB5
F1 score· 2020-08-09
92.27
best: 93.52 (Attach-Juxtapose Parser + BERT)
SOTA
Fast and Accurate Neural CRF Constituency Parsing arXiv:2008.03736
Constituency ParsingonCTB7
F1 score· 2020-08-09
91.55
best: 91.92 (CRF Parser + Electra)
Fast and Accurate Neural CRF Constituency Parsing arXiv:2008.03736
Constituency ParsingonPenn Treebank
F1 score· 2020-08-09
95.69
best: 96.43 (Hashing + XLNet)
Fast and Accurate Neural CRF Constituency Parsing arXiv:2008.03736