MIT Press

Table 9

Evaluation results for the fine-tuned LongFormer model on the manually annotated Wikipedia dataset. A generic neural NER model was also used for comparison.

	R_{di +qi}	ER_di	ER_qi	P_{di +qi}	WP_{di +qi}
RoBERTa (fined-tuned on Ontonotes)	0.845	0.810	0.801	0.770	0.836
Longformer (fine-tuned on TAB)	0.952	0.999	0.923	0.708	0.765

	R_{di +qi}	ER_di	ER_qi	P_{di +qi}	WP_{di +qi}
RoBERTa (fined-tuned on Ontonotes)	0.845	0.810	0.801	0.770	0.836
Longformer (fine-tuned on TAB)	0.952	0.999	0.923	0.708	0.765

View Large

This Feature Is Available To Subscribers Only