Relation Extraction On Ace 2004
Metrics
Cross Sentence
NER Micro F1
RE+ Micro F1
Results
Performance results of various models on this benchmark
| Paper Title | ||||
|---|---|---|---|---|
| Multi-turn QA | No | 83.6 | 49.4 | Entity-Relation Extraction as Multi-Turn Question Answering |
| PFN | No | 89.3 | 62.5 | A Partition Filter Network for Joint Entity and Relation Extraction |
| Ours: cross-sentence ALB | Yes | 90.3 | 62.2 | A Frustratingly Easy Approach for Entity and Relation Extraction |
| PL-Marker | Yes | 90.4 | 66.5 | Packed Levitated Marker for Entity and Relation Extraction |
| Joint w/ Global | No | 79.7 | 45.3 | - |
| SPTree | No | 81.8 | 48.4 | End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures |
| Attention | No | 79.6 | 45.7 | Going out on a limb: Joint Extraction of Entity Mentions and Relations without Dependency Trees |
| Table-Sequence | No | 88.6 | 59.6 | Two are Better than One: Joint Entity and Relation Extraction with Table-Sequence Encoders |
| multi-head + AT | No | 81.64 | 47.45 | Adversarial training for multi-context joint entity and relation extraction |
| multi-head | No | 81.16 | 47.14 | Joint entity recognition and relation extraction as a multi-head selection problem |
| DyGIE | Yes | 87.4 | - | A General Framework for Information Extraction using Dynamic Span Graphs |
0 of 11 row(s) selected.