Didactically Adaptive Learning Path Logics in LLM-Based Personalized Education Systems

Thesis (BA/MA/GR)

Advisor(s): Alisa Mehler(alisa.mehler(at)tum.de)

CONTEXT

Current LLM-based learning systems can generate personalized learning paths based on user parameters such as professional role or available time. However, true adaptivity in education requires more than parameter-based personalization: learning decisions must be didactically grounded, explainable, and pedagogically justified.

This thesis investigates how learning path generation can be enhanced through formal didactic models, competency-based adaptation, and transparent decision logic.

Rather than focusing on what an LLM can generate, the core research question is how adaptive learning behavior can be systematically designed, explained, and evaluated.

The goal is to develop and validate adaptive learning path logic that adjusts learning depth, sequencing, repetition, and competence progression in a scientifically grounded way, improving both learning quality and learner acceptance.

 

EXEMPLARY QUESTIONS:

Possible guiding research questions include:

• How can learner progress and competency development be reflected in adaptive learning paths?
• Which didactic models are suitable for structuring personalized learning journeys?
• How can learning path decisions be justified and made transparent to learners?
• What methods strengthen learner engagement through adaptive sequencing and repetition?
• How can adaptive systems balance personalization with pedagogical consistency?

Such questions can provide a strong basis for scientific research in this thesis context.

 

TASK(S)

• Review didactic frameworks for adaptive learning systems
• Formalize learning path decision logic and competence progression
• Integrate learner feedback and performance signals
• Implement dynamic adaptations (depth, sequence, repetition)
• Generate explanations that justify learning decisions
• Evaluate learning effectiveness and acceptance

 

REQUIREMENTS

• Strong interest in AI, information management, and adaptive systems
• High degree of autonomy and individual responsibility
• Experience with Python and prototypical implementation (depending on research focus)
• Interest in conducting scientific studies and analyzing qualitative and quantitative data 

• Structured, reliable, and self-motivated work style

   

FURTHER INFORMATION

The thesis can be written in English or German. The topic can also be adapted to your interests. If you have further questions, please do not hesitate to contact us directly. Please send your application including our specific application form, a current transcript of records and your CV to alisa.mehler(at)tum.de. Please also note that we can only consider applications with complete documents.