MAE (Valence): 0.168MAE (Arousal): 0.165MAE (Expectancy): 0.175MAE (Power): 7.9
Papers with Code · CC BY-SA 4.0
Emotion Recognition in Conversation
Given the transcript of a conversation along with speaker information of each constituent utterance, the ERC task aims to identify the emotion of each utterance from several pre-defined emotions. Formally, given the input sequence of N number of utterances [(u1, p1), (u2, p2), . . . , (uN , pN )], where each utterance ui = [ui,1, ui,2, . . . , ui,T ] consists of T words ui,j and spoken by party pi, the task is to predict the emotion label ei of each utterance ui. .
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.
데이터셋
16개논문
Accuracy of Sentiment: 48.62Macro-F1 of Sentiment: 30.33
Accuracy of Sentiment: 48.96Macro-F1 of Sentiment: 45.18
Accuracy of Sentiment: 49.65Macro-F1 of Sentiment: 45.40
Accuracy of Sentiment: 48.90Macro-F1 of Sentiment: 34.37
Accuracy of Sentiment: 51.50Macro-F1 of Sentiment: 48.02
MAE (Valence): 0.157MAE (Arousal): 0.161MAE (Expectancy): 0.168MAE (Power): 7.68
Accuracy of Sentiment: 51.24Macro-F1 of Sentiment: 46.96
Accuracy of Sentiment: 48.09Macro-F1 of Sentiment: 44.60
MAE (Valence): 0.16MAE (Arousal): 0.16MAE (Expectancy): 0.16MAE (Power): 7.70
Accuracy of Sentiment: 49.13Macro-F1 of Sentiment: 37.95
라이브러리
Papers with Code 리더보드 row에 연결된 구현 저장소입니다.