no code implementations • 27 Nov 2023 • Stephen MacNeil, Paul Denny, Andrew Tran, Juho Leinonen, Seth Bernstein, Arto Hellas, Sami Sarsa, Joanne Kim
Unlike syntax errors, for which a compiler or interpreter can issue a message, logic errors can be subtle.
no code implementations • 18 Jun 2023 • Paul Denny, Hassan Khosravi, Arto Hellas, Juho Leinonen, Sami Sarsa
In this study, we investigated the potential for LLMs to produce learning resources in an introductory programming context, by comparing the quality of the resources generated by an LLM with those created by students as part of a learnersourcing activity.
no code implementations • 9 Jun 2023 • Arto Hellas, Juho Leinonen, Sami Sarsa, Charles Koutcheme, Lilja Kujanpää, Juha Sorva
At the same time, the results highlight the unreliability of LLMs: LLMs make some of the same mistakes that students do, perhaps especially when formatting output as required by automated assessment systems.
no code implementations • 5 Jun 2023 • Paul Denny, James Prather, Brett A. Becker, James Finnie-Ansley, Arto Hellas, Juho Leinonen, Andrew Luxton-Reilly, Brent N. Reeves, Eddie Antonio Santos, Sami Sarsa
The computing education community has a rich history of pedagogical innovation designed to support students in introductory courses, and to support teachers in facilitating student learning.
no code implementations • 8 Apr 2023 • Juho Leinonen, Paul Denny, Stephen MacNeil, Sami Sarsa, Seth Bernstein, Joanne Kim, Andrew Tran, Arto Hellas
In this paper, we explore the potential of LLMs in generating explanations that can serve as examples to scaffold students' ability to understand and explain code.
no code implementations • 20 Oct 2022 • Juho Leinonen, Arto Hellas, Sami Sarsa, Brent Reeves, Paul Denny, James Prather, Brett A. Becker
Large language models can be used to create useful and novice-friendly enhancements to programming error messages that sometimes surpass the original programming error messages in interpretability and actionability.
no code implementations • 3 Jun 2022 • Sami Sarsa, Paul Denny, Arto Hellas, Juho Leinonen
Our analysis suggests that there is significant value in massive generative machine learning models as a tool for instructors, although there remains a need for some oversight to ensure the quality of the generated content before it is delivered to students.
no code implementations • 30 Dec 2021 • Sami Sarsa, Juho Leinonen, Arto Hellas
To evaluate how different aspects of DLKT models influence model performance, we test input and output layer variations found in the compared models that are independent of the main architectures.