Advances in Human-Computer Interaction with Virtual Agents and Large Language Models

The field of human-computer interaction is witnessing significant advancements with the integration of virtual agents and large language models (LLMs). Researchers are exploring innovative approaches to create more immersive and interactive digital humans, enabling them to simulate human-like personalities and emotions. This has far-reaching implications for various applications, including education, therapy, and gaming. Moreover, LLMs are being utilized to predict survey responses, generate personas, and enhance social media interactions. However, challenges such as biases, latency issues, and the lack of standardized evaluation frameworks need to be addressed. Noteworthy papers in this area include those that propose novel frameworks for persona generation, context-aware transcription interfaces, and deterministic AI agent personality expression. Notable papers in this area are: Integrating Personality into Digital Humans, which reviews LLM-driven approaches for creating digital humans with nuanced personality traits. LLM Generated Persona is a Promise with a Catch, which highlights the need for methodological rigor in persona generation using LLMs. Deterministic AI Agent Personality Expression through Standard Psychological Diagnostics, which demonstrates the ability to express deterministic and consistent personalities in AI agents using established psychological frameworks.

Sources

Integrating Personality into Digital Humans: A Review of LLM-Driven Approaches for Virtual Reality

Llms, Virtual Users, and Bias: Predicting Any Survey Question Without Human Data

Virtual Reality in Social Media: A New Era of Immersive Social Interactions

LLM Generated Persona is a Promise with a Catch

Since U Been Gone: Augmenting Context-Aware Transcriptions for Re-engaging in Immersive VR Meetings

Deterministic AI Agent Personality Expression through Standard Psychological Diagnostics

Your voice is your voice: Supporting Self-expression through Speech Generation and LLMs in Augmented and Alternative Communication

Enhancing Persona Consistency for LLMs' Role-Playing using Persona-Aware Contrastive Learning

Inducing Personality in LLM-Based Honeypot Agents: Measuring the Effect on Human-Like Agenda Generation

Built with on top of