Teaching
Type:
Upcoming Courses
- Neural Networks
- Seminar Understanding Large Language Models
Current Thesis Supervision
- Mechanistic Interpretability of Vision-Language Models (Topic)
- Explainable Reinforcement Learning (Topic)
- Explainable Speech Processing (Topic)
- Methods for Human-Robot Interaction by Distinguishing and Clarifying Uncertainty, Vagueness, and Ambiguity (Tentative Title)
- On the Ranking Behavior of Large Language Models in Numerical Comparisons
- Continual Disfluency Marker Training: Improving Performance on Impaired and Disfluent Speech
- Interpreting Looped Transformers for Systematic Generalization
2025
Courses
Thesis Supervision
- Methods for Downstream Continual Learning with Foundation Models
- Interpretability in Large Language Models: Explaining the Creation of Harmful Content Using Feature Attribution Methods
- Continual Learning in Panoptic Segmentation with Mixture of Expert Adapters
- Parameter-Efficient Fine-Tuning Methods as a Mixture of Experts for Continual Learning
- Expert Specialization in Mixture of Expert Models
2024
Courses
- Neural Networks
- LLM+XAI Reading Group
- Bio-Inspired AI Seminar
- WISDUM Reading Group
Thesis Supervision
- Active Vision for Embodied Agents Using Reinforcement Learning
- Length Generalization on Multi-Digit Integer Addition with Transformers
- Concept-Based Explanation for Continual Learning Models
- LeoLM: Evaluating and Enhancing German Language Proficiency in Pretrained Large Language Models
2023
Courses
- Neural Networks
- Neural Networks Seminar
- Bio-Inspired AI Seminar
- WISDUM Reading Group
Thesis Supervision
- Continual Learning for Language-Conditioned Robotic Manipulation
- Multivariate Normal Methods in Pre-trained Models for Continual Learning
- Generalization of Transformer-Based Models on Visual Question Answering Tasks
- Improving Compositional Generalization By Learning Concepts Individually
2022
Courses
- Neural Networks
- Neural Networks Seminar
- Bio-Inspired AI Seminar
- WISDUM Reading Group
Thesis Supervision
- Learning Concepts a Developmental Lifelong Learning Approach to Visual Question Answering
- Benchmarking Faithfulness: Towards Accurate Natural Language Explanations in Vision-Language Tasks
2021
Courses
- Neural Networks Seminar
- Bio-Inspired AI Seminar
- WISDUM Reading Group
Thesis Supervision
- Tackling The Binding Problem And Compositional Generalization In Multimodal Language Learning
- Learning Bidirectional Translation Between Robot Actions and Linguistic Descriptions
2020
Courses
- Neural Networks Seminar
- Bio-Inspired AI Seminar
- WISDUM Reading Group
Thesis Supervision
- Using the Reformer for Efficient Summarization
- Generalization in Multi-Modal Language Learning from Simulation
- Commonsense Validation and Explanation
2015
Courses
- AI Reading Group
2014
Courses
- Spatial Reasoning Tea
2013
Courses
- Spatial Reasoning Tea
2012
Courses
- Spatial Reasoning Tea
2008
Courses
- Mathematics III
2007
Courses
- Linear Algebra II
- Linear Algebra I + II Seminar
2006
Courses
- Linear Algebra I