Year: 2024
Type: article
Abstract: We propose a way of mitigating generative AI’s (GAI) inherent untrustworthiness by exploiting its ability to generate multiple versions of code and tests, facilitating comparative analysis across versions. Instead of relying on the quality of a single test or code module, this “Differential GAI” approach promotes more reliable quality evaluation through version diversity.
Source: IEEE Software
Authors: Marcus Kessel, Colin Atkinson
Institution: University of Mannheim
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4402916917