FAMILIARITY OF THE UNIVERSITY STUDENTS OF LANDSCAPE ENGINEERING WITH LARGE LANGUAGE MODELS

Abstract

The recent advances in machine learning led to exponential improvement in the field of artificial intelligence and large language models (LLMs). For university study and university students, LLMs can offer opportunities in several areas. But on the other hand, output from LLMs should be handled using critical thinking. In our research, we wanted to investigate the level of awareness of publicly available LLMs among the bachelor and master students (n=22) of Landscape Engineering program at the Slovak University of Agriculture in Nitra and the purpose of LLMs use in Slovak language. All the respondents heard about ChatGPT as an example of LLMs. The awareness of other LLMs was lower as expected, in the decreasing order of Gemini, Copilot and Perplexity. Majority of students found the AI tools at least �somewhat useful� with their daily tasks. More than 10% of the students from both groups concluded that AI tools provide substantial support for their education. According to our results, students use generally LLMs mostly for writing parts of assignments and projects (74.3%) and AI powered web search services (63.6%). In both cases more than 50% of respondents found LLMs useful for these tasks. In general, bachelor group showed to be more experienced in using LLMs for different purposes. In contrast to master group, bachelor group found LLMs very useful for explaining topics that they could not understand using LLMs as a tutor (nearly 70%) as well as for brainstorming and exploring new ideas (46%). Especially these last two usage areas can provide considerable support to traditional classroom education.

Ähnliche Arbeiten