机器人
计算机科学
人机交互
谈判
任务(项目管理)
背景(考古学)
人机交互
审议
社交机器人
人工智能
移动机器人
工程类
机器人控制
政治学
政治
古生物学
法学
生物
系统工程
作者
Callie Y. Kim,Christine P. Lee,Bilge Mutlu
标识
DOI:10.1145/3610977.3634966
摘要
Large-language models (LLMs) hold significant promise in improving\nhuman-robot interaction, offering advanced conversational skills and\nversatility in managing diverse, open-ended user requests in various tasks and\ndomains. Despite the potential to transform human-robot interaction, very\nlittle is known about the distinctive design requirements for utilizing LLMs in\nrobots, which may differ from text and voice interaction and vary by task and\ncontext. To better understand these requirements, we conducted a user study (n\n= 32) comparing an LLM-powered social robot against text- and voice-based\nagents, analyzing task-based requirements in conversational tasks, including\nchoose, generate, execute, and negotiate. Our findings show that LLM-powered\nrobots elevate expectations for sophisticated non-verbal cues and excel in\nconnection-building and deliberation, but fall short in logical communication\nand may induce anxiety. We provide design implications both for robots\nintegrating LLMs and for fine-tuning LLMs for use with robots.\n
科研通智能强力驱动
Strongly Powered by AbleSci AI