2024年人机情感交互领域热点回眸  

Review of research hotspots in affective human computer interaction in 2024

作　　者：谷雨 任福继 GU Yu;REN Fuji(School of Computer Science and Engineering,University of Electronic Science and Technology of China,Chengdu 611731,China)

机构地区：[1]电子科技大学计算机科学与工程学院,成都611731

出　　处：《科技导报》2025年第1期132-142,共11页Science & Technology Review

基　　金：国家自然科学基金联合重点项目(U24A20250)。

摘　　要：“人机情感交互”研究旨在创建一种能感知、识别和理解人的情感,并能针对人的情感作出智慧、灵敏、自然回应的智能系统,是实现机器人自然化、拟人化、人格化的基础技术和重要前提。2024年,大语言模型的进步给该领域带来了深刻的变革。从交互理论、情感认知、情感调节及交互效果评估层面回眸该领域年度研究进展,并探讨在情感认知可计算模型、情感调控可进化交互环路、情感交互可解释评估体系等方面存在的理论与技术问题。指出了人机情感交互领域未来研究主要方向为情感认知可计算建模方法、情感调控可进化交互环路、情感交互可解释评估体系等。Research on affective human-machine interaction aims to create an intelligent system capable of perceiving,recognizing,and understanding human emotions,and responding with wisdom,sensitivity,and naturalness.It serves as a fundamental technology and a crucial prerequisite for achieving naturalization,anthropomorphism,and personalization in robotics.In 2024 advancements in large language models brought profound changes to this field.This article reviews the progress in this field in 2024 in terms of interaction theory,emotional cognition,emotional regulation,and evaluation of interaction effects.Additionally,it discusses the theoretical and technical challenges that exist in areas such as computational models of emotional cognition,evolutionary interaction loops for emotional regulation,and explainable evaluation systems for emotional interaction.

关 键 词：人机情感交互 多模态大模型 情感认知 情感调节 交互效果评估 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]