Temporal Relation Classification on TB-DENSE
68.1F-scoreUnified-Framework
Evaluation Results
| Method | Links | |
|---|---|---|
| Unified-FrameworkPretrained Model=RoBERTa-Large2023.05 | 68.1 | |
| TIMERSPretrained Model=BERT-Large2023.05 | 67.8 | |
| Syntax-guided Graph TransformerPre-trained Model=BERT Large, Gold event mentions as input=true2021.04 | 67.1 | |
| SyntacticPretrained Model=BERT-Large, additional training resources=true2023.05 | 67.1 | |
| DEERPre-trained Model=RoBERTa Large, Gold event mentions as input=true2021.04 | 66.8 | |
| DEERPretrained Model=RoBERTa-Large2023.05 | 66.8 | |
| ECONETProtocol=Best single model, Backbone=RoBERTa-Large2020.12 | 66.8 | |
| Syntax-guided Graph TransformerPre-trained Model=BERT Base, Gold event mentions as input=true2021.04 | 66.7 | |
| SyntacticPretrained Model=BERT-Base, additional training resources=true2023.05 | 66.7 | |
| Zhang et al. (2021)Protocol=Best single model2020.12 | 66.7 | |
| Unified-FrameworkPretrained Model=BERT-Base2023.05 | 66.4 | |
| Enhanced-BaselinePretrained Model=RoBERTa-Large2023.05 | 66.2 | |
| PSLPre-trained Model=RoBERTa Large, Gold event mentions as input=true2021.04 | 65.2 | |
| PSLPretrained Model=RoBERTa-Large2023.05 | 65.2 | |
| RoBERTa-Large + GeneratorProtocol=Average of 3 seeds2020.12 | 65.2 | |
| Enhanced-BaselinePretrained Model=BERT-Base2023.05 | 64.8 | |
| TacoLMProtocol=Average of 3 seeds2020.12 | 64.8 | |
| RoBERTa-Large + ECONETProtocol=Average of 3 seeds2020.12 | 64.8 | |
| HNP19Pre-trained Model=BERT Base, Gold event mentions as input=true2021.04 | 64.5 | |
| HNPPretrained Model=BERT-Base2023.05 | 64.5 | |
| BERT-Large + ECONETProtocol=Average of 3 seeds2020.12 | 63 | |
| BERT-LargeProtocol=Average of 3 seeds2020.12 | 62.8 | |
| RoBERTa-LargeProtocol=Average of 3 seeds2020.12 | 62.8 | |
| LSTMPre-trained Model=BERT Base, Gold event mentions as input=true2021.04 | 62.2 | |
| LSTMPretrained Model=BERT-Base2023.05 | 62.2 |