How Do We Research Human-Robot Interaction in the Age of Large Language Models? A Systematic Review

A research team led by Yufeng Wang published a systematic review analyzing 86 studies on LLMs in Human-Robot Interaction (HRI). The review, following PRISMA guidelines, found LLMs are reshaping core HRI fundamentals like context sensing and social interaction generation. It highlights that current research is largely exploratory with fragmented methodologies, and provides key design guidelines for future work at the LLM-HRI intersection.