Table 6.

Experimental results on ACE05 English

Model	Feature	F1(%)
Maximum Entropy (Kambhatla 2004)	Multiple features	52.80
$Mix-CNN$ (Zheng et al. 2016)	Multi-grained local features	53.60
SVM (Zhou et al. 2005)	Phrase parsing	68.00
BERT (Wang and Lu 2020)	Table-sequence encoders	70.40
BERT (Zhong and Chen 2020)	Entity structure	73.10
Ours	Entity semantics & structure	77.70