Year: 2024
Type: conference-paper
Abstract: The rapid advancement of Large Language Models (LLMs) has opened new avenues for automating and enhancing the process of systematic literature reviews. This study investigates the effectiveness of three state-of-the-art LLMs — ChatGPT4o, LLaMA, and Gemini — in conducting literature reviews on the intersection of virtual reality and mental health, focusing on works by a renowned author in the field, Mel Slater. We defined two objective and two subjective questions to evaluate the performance of these models against a ground truth dataset. Our findings reveal significant insights into the accuracy, reliability, and limitations of each model, providing a comprehensive analysis of their potential and constraints. This study aims to guide future applications of LLMs in academic research, highlighting the transformative potential of these technologies in conducting systematic
Authors: Paulo Guedes, Pedro Abrantes, João Marcelo Teixeira, Veronica Teichrieb
Institution: Universidade Federal de Pernambuco
Language: English
Subfield: Statistics, Probability and Uncertainty
Field: Decision Sciences
Domain: Social Sciences
Open Access status: gold
OpenAlex ID: W4405030245