首页|期刊导航|Tsinghua Science and Technology|Multi-Affection Prompt Learning for Sentiment,Emotion,and Sarcasm Joint Detection in Conversations

Multi-Affection Prompt Learning for Sentiment,Emotion,and Sarcasm Joint Detection in ConversationsOA

中文摘要

As a new trend in Natural Language Processing(NLP),prompt tuning has been explored to provide a reliable answer without requiring massive labeled samples and training learning in sentiment and emotion detection tasks.However,how to effectively encode the commonness and uniquesness across difference affections into prompts sets a limit to the potential of multi-affection joint detection.To fill this gap,we propose a multi-affection prompt(MAP)learning framework that takes both the commonness of multiple affections and the uniqueness of specific affection into consideration.More specifically,two different prompt encoders are first proposed to elaborate the multi-task shared prompt and the task-specific prompt,respectively.Second,a multitask prompt interaction learning layer is proposed to capture the correlation between the multi-task and taskspecific prompts.MAP adopts separate multi-task and task-specific prompts to learn different vectors for different affection tasks,thus mitigating the affection discrepancy of the[MAsK]token in the masked language modeling task.Extensive experiments on two benchmark datasets show that our proposed method can significantly improve the multi-task generalization capability of PLMs,and yield better results than other stateof-the-art(SOTA)baselines,by the margin of 2.7%and 3.4%.

Yazhou Zhang;Yang Yu;Xiangkai Wang;Xiang Li;Hui Liang;Prayag Tiwari

College of Intelligence and Computing,Tianjin University,Tianjin 300350,ChinaSchool of Information Science and Engineering,Lanzhou University,Lanzhou 730000,ChinaShandong Zhengyun Information Technology Limited Company,Jinan 250353,ChinaQilu University of Technology(Shandong Academy of Sciences),Jinan 250353,ChinaZhengzhou University of Light Industry,Zhengzhou 450002,ChinaSchool of Information Technology,Halmstad University,Halmstad 30118,Sweden

信息技术与安全科学

sentiment analysisaffection detectionprompt learningmulti-task learning

《Tsinghua Science and Technology》 2026 (3)

P.1819-1837,19

supported by Hong Kong RGC Theme-based Research Scheme(No.T45-401/22-N)the National Natural Science Foundation of China(No.62006212)Fellowship from the China Postdoctoral Science Foundation(No.2023M733907)Natural Science Foundation of Hunan Province of China(No.242300421412)Foundation of Key Laboratory of Dependable Service Computing in Cyber-Physical-Society(Ministry of Education),and Chongqing University(No.CPSDSC202103).

10.26599/TST.2024.9010196

评论