SUPR
Impact of prompt programming on LLM outcome
Dnr:

NAISS 2025/22-526

Type:

NAISS Small Compute

Principal Investigator:

Ranim Khojah

Affiliation:

Chalmers tekniska högskola

Start Date:

2025-04-03

End Date:

2025-12-01

Primary Classification:

10205: Software Engineering

Webpage:

Allocation

Abstract

Currently there is a lot of work on prompt engineering and how it improves the outcome of large language models. I have recently worked on a small project with preliminary results that show that some prompt techniques do not have a dramatic impact on code generation. Now, I want to investigate this in more details and see how the choice of the LLM and other factors can contribute to the impact.