The rise of Artificial Intelligence (AI) and Generative AI (GenAI) marks a significant milestone in technological development, reshaping the way we approach complex challenges. These innovative technologies are redefining the boundaries of automation, efficiency, and personalized solutions in fields as diverse as healthcare, finance, and the creative arts. Their ability to process and analyze vast amounts of data, generate new content, and learn from interactions is revolutionizing traditional methods leading to a surge of exciting new opportunities. This transformative impact of AI and GenAI is not confined to commercial or creative sectors alone, it also extends to the field of education. As these technologies continue to evolve, they reveal their potential to fundamentally reshape how educational content is created, delivered, and interacted with.
The integration of AI and GenAI offers various opportunities in the field of teaching and education. AI and Explainable AI (XAI) methods can be leveraged for data-based grade predictions, revealing insights into one's own learning progress as well as identifying the most valuable next learning steps. Furthermore, Generative AI enables the development of tailored learning tools, such as interactive chatbots, and the creation of customized learning materials, particularly aimed at improving performance in areas where students may be struggling.
Scope of the thesis:
We at the ITEM aim to close the gap between sophisticated academia and tangible practical insights. While building on profound research, your thesis should generate actionable insights with practical applicability. Based on desk research, reviewing scientific literature, and systematic investigations we offer multiple theses with different thematic focuses.
Specifically, our project currently revolves around questions such as:
- How can Conversational AIs (ChatGPT, Bard) be leveraged to create different types of exercises (MC, open questions, calculation tasks)
- How can techniques such as LLM chaining, Chain of Verification or Retrieval Augmented Generation be used to reduce hallucination and reliably generate meaningful exercises and corresponding solutions for subsequent automation?
- How can Conversational AIs (ChatGPT, Bard) be trained on our own data (e.g., custom GPTs)?
- How can projects such as Lamini, Ollama or PrivateGPT be leveraged to achieve our goals with open-source language models?
Your profile:
- Strong interest and affinity for AI and especially GenAI
- Depending on the topic, initial programming knowledge would be an advantage but not required
- Organized, conscientious, and result-oriented working attitude
- Very good communication and writing skills in English
We offer:
- Practice-oriented research with high relevance and impact
- Methodological support
- Working at the intersection of business and technology
- A young, dynamic, and interdisciplinary team
Whom to contact?
If you find the topic interesting, please do not hesitate to contact Maximilian May (maximilian.may@unisg.ch). Please attach a short letter of motivation for this topic, your CV, and your current grade transcript to your e-mail.
If you find the topic interesting, please do not hesitate to contact Maximilian May (maximilian.may@unisg.ch). Please attach a short letter of motivation for this topic, your CV, and your current grade transcript to your e-mail.