Relation Extraction on revised TACRED (test)
84.1Micro-F1KLG
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| KLGCategory=Ours2022.12 | 84.1 | — | |
| KLG w/o DSCategory=Ours, Feature=without Dynamic k-Selection2022.12 | 83.6 | — | |
| Base Model(LS)Category=Ours, Feature=Label Smoothing2022.12 | 83.5 | — | |
| RECENTCategory=Classification-based2022.12 | 83.4 | — | |
| IRECategory=Classification-based2022.12 | 83.2 | — | |
| Base ModelCategory=Ours2022.12 | 83.1 | — | |
| Base Model(P)Category=Ours, Feature=Prompt P2022.12 | 82.7 | — | |
| KnowPromptCategory=Prompt-based2022.12 | 82.4 | — | |
| BARTCategory=Sequence-to-Sequence2022.12 | 81.5 | — | |
| PTRCategory=Prompt-based2022.12 | 81.4 | — | |
| LUKECategory=Classification-based2022.12 | 80.8 | — | |
| BOOTLEG ModelBackbone=SpanBERT2020.10 | — | 80.3 | |
| FINE-TUNINGExtra Data=w/o, Backbone=RoBERTa, Learning Paradigm=Fine-tuning2021.04 | — | 84.9 | |
| KnowBERT2020.10 | — | 79.3 | |
| KNOWBERTExtra Data=w/, Learning Paradigm=Fine-tuning2021.04 | — | 89.1 | |
| KnowPromptExtra Data=w/o, Backbone=RoBERTa, Learning Paradigm=Prompt-tuning2021.04 | — | 91.3 | |
| PTRExtra Data=w/o, Backbone=RoBERTa, Learning Paradigm=Prompt-tuning2021.04 | — | 90.9 | |
| SpanBERT2020.10 | — | 78 | |
| SPANBERTExtra Data=w/, Learning Paradigm=Fine-tuning2021.04 | — | 85.3 |