Role F1 (Preps): 72.4Function F1 (Preps): 82.8Full F1 (Preps): 71.6Tags (Full) Acc: 82.5
Papers with Code · CC BY-SA 4.0
Natural Language Understanding
Natural Language Understanding is an important field of Natural Language Processing which contains various tasks such as text classification, natural language inference and story comprehension. Applications enabled by natural language understanding range from question answering to automated reasoning.
Source: Find a Reasonable Ending for Stories: Does Logic Relation Help the Story Cloze Test?
개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.
벤치마크
Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.
Dataset기록된 모델 예시MetricPaperCode
BERT (pred POS/lemmas)
Role F1 (Preps): 72.4Function F1 (Preps): 82.8Full F1 (Preps): 71.6Tags (Full) Acc: 82.5
논문
코드
BERT
ECtHR Task A: 71.4 / 64.0ECtHR Task B: 87.6 / 77.8SCOTUS: 70.5 / 60.9EUR-LEX: 71.6 / 55.6
논문
코드
ConvBERT-DG
Average: 74.6Banking77 (Acc): 84.84CLINC150 (Acc): 93.53HWU64 (Acc): 86.71
-
-
ConvBERT + Pre + Multi
Average: 86.89Banking77 (Acc): 93.44CLINC150 (Acc): 92.38HWU64 (Acc): 97.11
-
-
데이터셋
6개논문
Role F1 (Preps): 62.2Function F1 (Preps): 73.4Full F1 (Preps): 58.9
Average: 82.1
Average: 89.9
라이브러리
Papers with Code 리더보드 row에 연결된 구현 저장소입니다.
github.com/1wy/bert
1 논문