Enhancing Multi-Person Dialogue with Large Language Models: A Structured Approach to Natural Communication
Takumi Murogaki, Toshikazu Nishimura · 2024
The rise of social networking services has increased text-based communication, often leading to misunderstandings. This study aims to develop a system using large language models (LLMs) like ChatGPT to provide real-time support in human dialogues. Traditional LLM chatbots, designed for one-on-one interactions, struggle with multi-person conversations, often leading to unnatural responses. This research proposes methods to enhance LLM's ability to distinguish between different speakers and improve its reasoning capabilities. By implementing "conversation structure tags" and simulating multi-person arguments, the system aims to generate natural, context-aware responses, enhancing the dialogue's quality and engagement.