Semantic Role Labeling on CoNLL 2012 (dev)
87.36F1LISA
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| LISAEmbeddings=ELMo, External Parse=Gold (Synthetic Upper Bound)2018.04 | 87.36 | 88.22 | 86.53 | — | |
| LISAEmbeddings=GloVe, External Parse=Gold (Synthetic Upper Bound)2018.04 | 86.43 | 87.57 | 85.32 | — | |
| DEEPATT (FFN, Ensemble)Nonlinearity=FFN, model_type=Ensemble2017.12 | 84.1 | 83.6 | 84.7 | 68.7 | |
| He et al. (2017)model_type=Ensemble2017.12 | 83.4 | 83.5 | 83.2 | 67.5 | |
| LISAEmbeddings=ELMo2018.04 | 83.37 | 84.19 | 82.56 | — | |
| LISAEmbeddings=ELMo, External Parse=D&M2018.04 | 83.36 | 84.09 | 82.65 | — | |
| SAEmbeddings=ELMo2018.04 | 83.23 | 84.35 | 82.14 | — | |
| Span-based SRL (Ours)Embeddings=ELMo, Evaluation Protocol=End-to-End2018.05 | 83 | — | — | — | |
| He et al.Embeddings=ELMo2018.04 | 83 | 82.1 | 84 | — | |
| DEEPATT (FFN)Nonlinearity=FFN, model_type=Single2017.12 | 82.9 | 82.2 | 83.6 | 66.7 | |
| LISAEmbeddings=GloVe, External Parse=D&M2018.04 | 82.05 | 82.97 | 81.14 | — | |
| DEEPATT (RNN)Nonlinearity=RNN, model_type=Single2017.12 | 81.6 | 81 | 82.3 | 64.6 | |
| He et al. (2017)model_type=Single2017.12 | 81.5 | 81.7 | 81.4 | 64.6 | |
| DEEPATT (CNN)Nonlinearity=CNN, model_type=Single2017.12 | 81.3 | 80.1 | 82.5 | 65 | |
| Zhou and Xu (2015)2017.12 | 81.1 | — | — | — | |
| SAEmbeddings=GloVe2018.04 | 81.02 | 82.32 | 79.76 | — | |
| LISAEmbeddings=GloVe2018.04 | 80.7 | 81.77 | 79.65 | — | |
| FitzGerald et al. (2015)model_type=Ensemble, variant=Struct.2017.12 | 79.7 | 81 | 78.5 | 60.9 | |
| Span-based SRL (Ours)Embeddings=GloVe, Evaluation Protocol=End-to-End2018.05 | 79.4 | — | — | — | |
| He et al.Embeddings=GloVe2018.04 | 79.4 | 79.2 | 79.7 | — | |
| Täckström et al. (2015)model_type=Ensemble, variant=Struct.2017.12 | 79.1 | 80.5 | 77.8 | 60.1 | |
| He et al. (2017)Model Variant=Product of Experts (POE), Evaluation Protocol=End-to-End2018.05 | 77.2 | — | — | — | |
| He et al. (2017)Evaluation Protocol=End-to-End2018.05 | 75.5 | — | — | — |