multimodal interaction
多模态交互：结合多种输入与输出通道（如语音、注视、手势、触摸、视觉叠加等）的交互方式，让人与计算系统的沟通更自然、灵活。
English Definition
"Interaction that combines multiple input and/or output modalities (e.g., speech, gaze, gesture, touch, visual overlays), enabling more natural and flexible communication with computational systems."
用法说明 · 针对中文母语者
在 HCI 中是相对于单模态（如纯文本或纯 GUI）的设计范式，不要与机器学习中的 multimodal learning 或 multimodal model 简单等同。
真实用例 · 3 条
"Supporting humans and AI in fluidly establishing, shifting, and maintaining shared attention through multimodal signals such as gaze, gestures, and AR highlights."
"We instantiate this framework in an AR-based prototype equipped with a real-time multimodal pipeline."
"Multimodal expressions, including natural language, gaze, and gestures, offered unique contributions to the establishment of common ground."