Relation Extraction On Redocred
평가 지표
F1
Ign F1
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | F1 | Ign F1 | Paper Title | Repository |
|---|---|---|---|---|
| DREEAM | 80.73 | 79.66 | DREEAM: Guiding Attention with Evidence for Improving Document-Level Relation Extraction | - |
| DocuNET | 77.87 | 77.26 | Document-level Relation Extraction as Semantic Segmentation | - |
| JEREX | 72.57 | 71.45 | An End-to-end Model for Entity-level Relation Extraction using Multi-instance Learning | - |
| ATLOP | 77.56 | 76.82 | Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling | - |
| SSR-PU | 78.86 | 77.67 | A Unified Positive-Unlabeled Learning Framework for Document-Level Relation Extraction with Different Levels of Labeling | |
| KD-DocRE | 78.28 | 77.60 | Document-Level Relation Extraction with Adaptive Focal Loss and Knowledge Distillation | - |
| HingeABL | 79.79 | 78.82 | Adaptive Hinge Balance Loss for Document-Level Relation Extraction |
0 of 7 row(s) selected.