Scholarlink

In the field of artificial intelligence, natural language processing is a technology that is advancing and being widely applied to recognize natural human language. This research will facilitate the development of natural interactions between humans and machines, in which machines will acquire the ability to comprehend human language, automatically adjust, and engage in proactive behaviour. In this paper, we investigate a virtual assistant that can proactively communicate and interact with users using artificial intelligence technology. The virtual assistant is based on a large language model, VinaLLaMA-7B-chat, which is a large modern language model for Vietnamese. It is built on the LLaMA-2-7B model and is supported by the TRL, accelerate, and LangChain libraries. The data source is used to search and query the collected and preprocessed database in preparation for the model calibration process. The application of virtual assistants in the communication, education, and service industries is widespread, and they are continuously enhancing system optimization. These assistants are highly accurate.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Virtual Personal Assistant Application Based on a Large Language Model

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Tan-Tien Nguyen Thi,
  • Huu-Huy Ngo,
  • Hung Linh Le,
  • Kim-Son Nguyen,
  • Man Ba Tuyen,
  • Tran Xuan Thanh

摘要

In the field of artificial intelligence, natural language processing is a technology that is advancing and being widely applied to recognize natural human language. This research will facilitate the development of natural interactions between humans and machines, in which machines will acquire the ability to comprehend human language, automatically adjust, and engage in proactive behaviour. In this paper, we investigate a virtual assistant that can proactively communicate and interact with users using artificial intelligence technology. The virtual assistant is based on a large language model, VinaLLaMA-7B-chat, which is a large modern language model for Vietnamese. It is built on the LLaMA-2-7B model and is supported by the TRL, accelerate, and LangChain libraries. The data source is used to search and query the collected and preprocessed database in preparation for the model calibration process. The application of virtual assistants in the communication, education, and service industries is widespread, and they are continuously enhancing system optimization. These assistants are highly accurate.