Application of Retrieval-Augmented Generation for Legal Documents
Loading...
Date
2024
Authors
Маринич, Антон
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The objective of this work is to examine the power of RAG on improving the knowledge of the model about Ukrainian traffic rules, selected as a representative example of legal documents. This work is useful, because it is easier to use a model to answer the questions about traffic rules, instead of going through long legal documents. The other advantage is that the information on the Internet might be out of date, whereas it is easy to update the information in the documents by using RAG.
Description
Keywords
baseline GPT-3.5-turbo model, powerful LLMs, GPT-4, research on traffic rules, coursework