Span-based Semantic Role Labeling on CoNLL 2005 (Out-of-domain (Brown))

85.1F1 Score

Ours

Updated 4d ago

Evaluation Results

Method
Ours 2021.09	85.1	86.4	83.8
Ours 2021.09	84.7	85.9	83.5
Zhang et al. 2021.09	83.7	83.8	83.6
SynG2G-Tr 2021.04	83.21	83.93	82.5
Mohammadshahi and Henderson 2021.09	83.2	83.9	82.5
Xia et al. 2021.09	83.2	83.1	83.3
BERT-LSTM-large 2019.04	82	81.9	82.1
Shi and Lin 2021.09	82	81.9	82.1
Paolini et al. 2021.09	82	-	-
Jia et al. (2022) 2021.04	81.9	-	-
Jindal et al. (2020) 2021.04	81.44	81.52	81.36
Zhou et al. 2021.09	81.4	81.9	81
BERT-LSTM-base 2019.04	80.9	80.7	81.2
Marcheggiani and Titov 2021.09	80.6	80.5	80.7
SynG2G-Tr 2021.04	80.53	80.01	81.07
Li et al. (2019) 2019.04	80.5	80.6	80.4
Li et al. 2021.09	80.5	80.6	80.4
Li et al. (2019) 2021.04	80.5	80.6	80.4
He et al. (2018a) 2019.04	80.4	-	-
He et al. (2018a) 2021.04	80.4	-	-
Jindal et al. 2021.09	80.2	80.3	80.1
Zhang et al. (2021) 2021.04	80.2	80.3	80.1
Ouchi et al. (2018) 2019.04	79.6	81	78.4
Ouchi et al. (2018) 2019.04	78.7	79.9	77.5
Ouchi et al. (2018) 2021.04	78.7	79.9	77.5
Strubell et al. (2018)† 2021.04	78.25	79.02	77.49
SynG2G-Tr (w/o BERT) 2021.04	77.45	77.73	77.18
Strubell et al. (2018)† 2021.04	76.54	76.65	76.44
Strubell et al. (2018) 2019.04	76.5	76.7	76.4
Li et al. (2019) 2021.04	76.4	74.7	78.1
He et al. (2018a) 2021.04	76.1	73.9	78.4
Xia et al. (2020) 2021.04	75.86	76.3	75.42
SynG2G-Tr (w/o BERT) 2021.04	75.26	73.92	76.65
Strubell et al. (2018) 2021.04	74.66	75.8	73.54
Xia et al. (2020) 2021.04	74.19	73.47	74.92
Tan et al. (2017) 2021.04	74.1	73.5	74.6
He et al. (2018a) 2021.04	73.7	-	-
He et al. (2017) 2021.04	73.6	74.9	72.4
Ouchi et al. (2018) 2021.04	73.1	76	70.4
Xia et al. (2019) 2021.04	72.9	73.7	72
He et al. (2018a) 2021.04	70.8	69.7	71.9