Charles Koutcheme
Doctoral Researcher
Doctoral Researcher
T313 Dept. Computer Science
Doctoral Researcher
Department of Computer Science
I’m a doctoral researcher in the Computer Science department at Aalto University. My Ph.D. dissertation contributes methods to automate programming feedback using open-source language models, with a focus on supporting novice programmers in large-scale learning environments where individual feedback is hard to scale.
The broader goal of my research is to move beyond systems that rely on large language models and instead build lightweight, locally deployable learning tools powered by small models. My work blends reinforcement learning and source code modelling approaches with knowledge from programming education to effectively support students.
Full researcher profile
https://research.aalto.fi/...
E-post
charles.koutcheme@aalto.fi
Postadress
Konemiehentie 2
02150 Espoo
Finland
Kompetensområde
Natural Language Processing, Computing Education Research, Educational data mining, 113 Computer and information sciences
Forskningsgrupp
- Lecturer Haaranen Lassi group, Doctoral Researcher
Publikationer
Direct Repair Optimization : Training Small Language Models For Educational Program Repair Improves Feedback
Charlie Koutcheme, Nicola Dainese, Arto Hellas
2025
Proceedings of the 20th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2025)
Evaluating Language Models for Generating and Judging Programming Feedback
Charles Koutcheme, Nicola Dainese, Sami Sarsa, Arto Hellas, Juho Leinonen, Syed Ashraf, Paul Denny
2025
SIGCSE TS 2025 - Proceedings of the 56th ACM Technical Symposium on Computer Science Education
Reinforcement Learning for Programming Feedback : Aligning Small Language Models Without Human Preferences
Charlie Koutcheme, Nicola Dainese, Arto Hellas
2025
Proceedings of 9th Educational Data Mining in Computer Science Education (CSEDM) Workshop
AI Chatbots vs. Traditional Search : a Comparative Study on Student Information Retrieval
Leonardo Pasquarelli, Charles Koutcheme, Arto Hellas
2025
IEEE Frontiers in Education Conference
Benchmarking Educational Program Repair
Charlie Koutcheme, Nicola Dainese, Sami Sarsa, Juho Leinonen, Arto Hellas, Paul Denny
2024
Open Source Language Models Can Provide Feedback : Evaluating LLMs' Ability to Help Students Using GPT-4-As-A-Judge
Charles Koutcheme, Nicola Dainese, Sami Sarsa, Arto Hellas, Juho Leinonen, Paul Denny
2024
ITiCSE 2024 - Proceedings of the 2024 Conference Innovation and Technology in Computer Science Education
Propagating Large Language Models Programming Feedback
Charles Koutcheme, Arto Hellas
2024
L@S 2024 - Proceedings of the 11th ACM Conference on Learning @ Scale
Using Program Repair as a Proxy for Language Models’ Feedback Ability in Programming Education
Charles Koutcheme, Nicola Dainese, Arto Hellas
2024
Proceedings of the 19th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2024)
Let's Ask AI About Their Programs : Exploring ChatGPT's Answers To Program Comprehension Questions
Teemu Lehtinen, Charles Koutcheme, Arto Hellas
2024
Proceedings of the 46th International Conference on Software Engineering: Software Engineering Education and Training
Exploring the Responses of Large Language Models to Beginner Programmers’ Help Requests
Arto Hellas, Juho Leinonen, Sami Sarsa, Charles Koutcheme, Lilja Kujanpää, Juha Sorva
2023
ICER '23: Proceedings of the 2023 ACM Conference on International Computing Education Research - Volume 1
Automated Program Repair Using Generative Models for Code Infilling
Charles Koutcheme, Sami Sarsa, Juho Leinonen, Arto Hellas, Paul Denny
2023
Artificial Intelligence in Education : 24th International Conference, AIED 2023, Tokyo, Japan, July 3–7, 2023, Proceedings
Evaluating Distance Measures for Program Repair
Charles Koutcheme, Sami Sarsa, Juho Leinonen, Lassi Haaranen, Arto Hellas
2023
ICER '23: Proceedings of the 2023 ACM Conference on International Computing Education Research - Volume 1
Training Language Models for Programming Feedback Using Automated Repair Tools
Charles Koutcheme
2023
Artificial Intelligence in Education
Exploring How Students Solve Open-ended Assignments
Charles Koutcheme, Artturi Tilanterä, Aleksi Peltonen, Arto Hellas, Lassi Haaranen
2022
ITiCSE 2022 - Proceedings of the 27th ACM Conference on Innovation and Technology in Computer Science Education
Methodological Considerations for Predicting At-risk Students
Charles Koutcheme, Sami Sarsa, Arto Hellas, Lassi Haaranen, Juho Leinonen
2022
ACE '22: Australasian Computing Education Conference
Towards Open Natural Language Feedback Generation for Novice Programmers using Large Language Models
Charles Koutcheme
2022
Speeding Up Automated Assessment of Programming Exercises
Sami Sarsa, Juho Leinonen, Charles Koutcheme, Arto Hellas
2022
Proceedings of the 2022 Conference on United Kingdom & Ireland Computing Education Research
Analyzing Fine-Grained Material Usage Behavior
Charles Koutcheme, Juho Leinonen, Juha Sorva, Arto Hellas
2021
Seventh SPLICE Workshop at SIGCSE 2021 “CS Education Infrastructure for All III: From Ideas to Practice”