aalto1 untyped-item.component.html
Evaluating the readability of code generated by large language models
Loading...
Files
Aalto login required (access for Aalto Staff only).
URL
Journal Title
Journal ISSN
Volume Title
Perustieteiden korkeakoulu |
Bachelor's thesis
Electronic archive copy is available locally at the Harald Herlin Learning Centre. The staff of Aalto University has access to the electronic bachelor's theses by logging into Aaltodoc with their personal Aalto user ID. Read more about the availability of the bachelor's theses.
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Authors
Date
Department
Major/Subject
Mcode
SCI3095
Degree programme
Language
en
Pages
25
Series
Abstract
The recent development of large language models (LLMs) and artificial intelligence (AI) has stimulated the emergence of various AI-powered tools for programming. As developers and companies have started to utilize such tools in software development, numerous concerns have been raised regarding their ability to assist developers and improve efficiency.
This thesis focuses on code readability and its evaluation in the context of AI-generated code, providing a summary of the current research in this area. The literature review shows that studies on this topic typically involve code generation using various methods, including solving programming challenges, experimenting with human developers, and utilizing search queries. Following code generation, the readability evaluation is conducted through one or more approaches including human assessment, readability models, and metric-based analyses.
The results indicate that LLMs and AI demonstrate commendable performance in generating code with good readability. However, it is important to note the inherent limitations of each evaluation method, and readability alone does not guarantee that the generated code is immediately usable. This thesis further addresses the limitations of readability evaluation and additional concerns regarding AI-generated code.