Scarecrows in Oz: Large Language Models in HRI

Large Language Models (LLMs) have been the focus of intense interest in the past few years for the artificial intelligence (AI) community and their use in interactive robots for industry has had equal interest; however, there do not currently exist guidelines for or categorizations of their use in different application spaces. In this workshop, we bring together academic researchers and industry practitioners who are using or are interested in LLMs for human-robot interaction (HRI), and who can contribute to the development of a higher level, community-wide evaluation of how LLMs can fit correctly and defensibly into the future of HRI. Relevant topics to this workshop include HRI studies that involve LLMs directly or indirectly, technical contributions that involve utilizing or modifying LLMs, and HRI studies that utilize the idea of "Scarecrows'' within a larger system. We also intend to incorporate work on the broader questions of how these models should be conceptualized within the framework of effective, responsible HRI.