HyperAI超神经

Dependency Parsing On Penn Treebank

评估指标

LAS
POS
UAS

评测结果

各个模型在此基准测试上的表现结果

比较表格
模型名称LASPOSUAS
globally-normalized-transition-based-neural92.7997.4494.61
efficient-second-order-treecrf-for-neural94.49-96.14
deep-biaffine-attention-for-neural-dependency95.75-97.29
second-order-neural-dependency-parsing-with95.34-96.91
generalizing-natural-language-analysis-194.70-96.44
distilling-an-ensemble-of-greedy-dependency92.0697.4494.26
deep-biaffine-attention-for-neural-dependency94.22-95.87
graph-based-dependency-parsing-with-graph94.3197.395.97
rethinking-self-attention-an-interpretable96.2697.397.42
an-improved-neural-network-model-for-joint93.8797.9795.51
structured-training-for-neural-network92.0697.394.01
recursive-non-autoregressive-graph-to-graph95.01-96.66
head-driven-phrase-structure-grammar-parsing95.7297.397.20
semi-supervised-sequence-modeling-with-cross95.02-96.61
enhancing-structure-aware-encoder-with95.92-97.30
left-to-right-dependency-parsing-with-pointer94.43-96.04
simple-and-accurate-dependency-parsing-using91.997.4493.99
training-with-exploration-improves-a-greedy91.4297.393.56
stack-pointer-networks-for-dependency-parsing94.1997.395.87
simple-and-accurate-dependency-parsing-using91.097.393.1
automated-concatenation-of-embeddings-for-195.8-97.2
模型 2293.2-95.42