An LLM-based Knowledge Service System for Libraries Integrating the ReAct Model
Focusing on library knowledge service in the era of Large Language Models(LLMs),this article explores the way of developing an LLM-based knowledge service system that incorporates the ReAct model in libraries using the genealogy knowledge as the subject of experiment.By converting RDF data into natural language text,it creates a data set and uses the ReAct model to enable the LLM system to perform tasks such as data query,multilingual query,resource retrieval,resource recommendation and interpretation,which demonstrates the capability of the LLM integrated with the ReAct model in natural language processing and context understanding.However,this study also reveals the problems related to the understanding of multilingual questions,the consistency of fuzzy query results,and the provision of additional background knowledge.This article presents a new perspective on the innovative development of library knowledge services based on LLMs,and explores a new pattern for libraries to improve the effectiveness of knowledge services and personalized services by integrating the reasoning and action capabilities of LLM system under the ReAct model.
library knowledge servicelarge language modelAI 2.0 eraReAct model
NETL
NSTL
万方数据
