HyperAI초신경
홈뉴스최신 연구 논문튜토리얼데이터셋백과사전SOTALLM 모델GPU 랭킹컨퍼런스
전체 검색
소개
HyperAI초신경

Joint Entity And Relation Extraction On

평가 지표

Cross Sentence
Entity F1

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름
Cross Sentence
Entity F1
Paper TitleRepository
SpERTNo70.33Span-based Joint Entity and Relation Extraction with Transformer Pre-training
TriMF-70.17A Trigger-Sense Memory Flow Framework for Joint Entity and Relation Extraction
SciIENo64.20Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Construction
PL-MarkerYes69.9Packed Levitated Marker for Entity and Relation Extraction
SpERT.PL (SciBERT)No70.53Joint Entity and Relation Extraction from Scientific Documents: Role of Linguistic Information and Entity Types
DyGIEYes65.2A General Framework for Information Extraction using Dynamic Span Graphs
MGENo68.4A Multi-Gate Encoder for Joint Entity and Relation Extraction-
PFNNo66.8A Partition Filter Network for Joint Entity and Relation Extraction
Ours: cross-sentenceYes68.9A Frustratingly Easy Approach for Entity and Relation Extraction
SpERT (with overlap)No70.3Span-based Joint Entity and Relation Extraction with Transformer Pre-training
DyGIE++Yes67.50Entity, Relation, and Event Extraction with Contextualized Span Representations
0 of 11 row(s) selected.