The paper focuses on the potential benefits and challenges of integrating Large Language Models (LLMs) into Extended Reality (XR) environments. The authors argue that embedding LLMs into XR can lead to more inclusive and engaging experiences for users. One promising application is in creating more interactive and personalized virtual avatars or narratives. However, the paper also highlights potential privacy concerns, as the combination of user-provided information and biometric data could lead to new forms of privacy invasions. The research emphasizes the need to investigate these potential privacy issues in conjunction with user privacy concerns and preferences. Despite these challenges, the authors see the integration of LLMs into XR as a promising research area with several opportunities.
Publication date: 7 Feb 2024
Project Page: N/A
Paper: https://arxiv.org/pdf/2402.03907
