HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Open Information Extraction On Benchie

평가 지표

F1

Precision

Recall

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	F1	Precision	Recall	Paper Title	Repository
CompactIE	0.318	0.414	0.258	CompactIE: Compact Facts in Open Information Extraction
Stanford OIE	0.13	0.11	0.16	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation
OpenIE6	0.25	0.31	0.21	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation
ClausIE	0.34	0.50	0.26	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation
MinIE	-	0.43	0.28	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation
ROIE-T	0.13	0.37	0.08	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation
Naive OIE	0.03	0.03	0.02	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation
M2OIE (ZH)	0.17	0.26	0.13	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation
ROIE-N	0.13	0.20	0.09	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation
M2OIE (EN)	0.23	0.39	-	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation
M2OIE (DE)	0.04	0.09	0.03	BenchIE: A Framework for Multi-Faceted Fact-Based Open Information Extraction Evaluation

0 of 11 row(s) selected.