Teaching & Supervision

Supervised Theses

I supervise student projects in Natural Language Processing, focusing on fairness, content moderation, and transfer learning. Below is an overview of my past and current teaching and supervision:

• Spring 2026 – TU Munich
  • Master’s Thesis: Multilingual Harmful Content Detection via Agents Debate
  • Master’s Thesis: Expert Agents to Generalize Harmful Content Detection

  Note: I am currently not accepting additional thesis students.

• Winter 2025 – TU Munich
  • Bachelor’s Thesis: Fine-Grained Multi-Attribute Simulation of Subjective Task Annotations Using LLMs
  • Bachelor’s Thesis: Large Language Models on Persian Literature: A Category-Wise Evaluation of Exam Questions
  • Bachelor’s Thesis: Explaining Intermediate Training Success in Hate Speech Detection via Dataset Similarity
• Spring 2025 – TU Munich
  • Master’s Thesis: Mitigating Temporal Bias in Hate Speech Detection
  • Bachelor’s Thesis: Nationality Bias Detection in Generative LLMs: A Perplexity-Based Study

For details on our chair’s current thesis offerings, see the DAST Projects & Thesis Topics page.

Teaching Engagements