Scarecrows in Oz: Large Language Models in HRI
Open Access
- 11 March 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1338-1340
Abstract
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.Keywords
Funding Information
- Air Force Office of Scientific Research (FA9550-20-1-0089)
- Office of Naval Research (N00014-21-1-2418)
- National Science Foundation (IIS-2024878, IIS-2145642)
This publication has 14 references indexed in Scilit:
- Scarecrows in Oz: The Use of Large Language Models in HRIACM Transactions on Human-Robot Interaction, 2024
- Large Language Models as Zero-Shot Human Models for Human-Robot Interaction2023
- Voice in the Machine: Ethical Considerations for Language-Capable RobotsCommunications of the ACM, 2023
- Improved Trust in Human-Robot Collaboration With ChatGPTIEEE Access, 2023
- Grounding Language with Visual Affordances over Unstructured Data2023
- Data-Efficient Learning of Natural Language to Linear Temporal Logic Translators for Robot Task Specification2023
- Robots Enact Malignant Stereotypes2022
- Language Models of Code are Few-Shot Commonsense Learners2022
- On the Dangers of Stochastic Parrots2021
- Robots That Use LanguageAnnual Review of Control, Robotics, and Autonomous Systems, 2020