teaching

Bachelor and Master thesis supervision

  • ConceptExplorer: A Dynamic Frontend for Concept Visualization and Indexing with Django (with Yuxuan Zahed)
  • How Multilingual are Existing Multilingual Benchmark Datasets? A Systematic Investigation (with Jiahao Lu)
  • Breaking the Script Barrier: Improving Cross-Lingual Transfer in Multilingual Models (with Orgest Xhelili, published in EMNLP 2024)
  • Understanding In-Context Machine Translation for Low-Resource Languages: A Case Study on Manchu (with Renhao Pei, published in ACL 2025)
  • Detecting MultiLingual Relation Specific Neurons in LLM (with Runsheng Chen, published in Arxiv 2025)
  • Prioritized Training on Worth-Learning Samples with Your Pretrained Model (with Qi Feng, published in ACL SRW 2025)
  • Language-Specific Concepts in LLMs: Tracing Paths Through Latent Language (with Nicolas Hoffmann)

Colloquium und Repetitorium zu Computerlinguistisches Arbeiten (Colloquium and Repetitorium on Computational Linguistics)

SS 2023, 2024, Bachelor class

In this class we will talk about useful topics for the preparation of a Bachelor thesis. Some companies will briefly present the work of a computer linguists in a their company and students will have to chance to get an overview of the tasks and projects they will be working on as computational linguists.

Einführung in die Computerlinguistik (Introduction to Computational Linguistics)

WS 2023/2024, WS 2024/2025, Bachelor class

This class will be divided into two parts: the linguistics part and the statistics part. In the linguistic part, we will talk about topics such as Phonetics, Morphology, and Syntax. In the statistics part, we will talk about topics such as Naive Bayes, Machine Translation, and Parsing. The class aims to provide a foundational understanding of computational linguistics.

Deep Learning for Natural Language Processing

WS 2024/2025, Master class

In this class, we will talk about the fundamental knowledge of deep learning and how this knowledge is applied in natural language processing. This class offers a theoretical part and a practical part. In the practical part, students will be given 5 projects from which the students can obtain some hands-on experience in how to use a deep learning framework (PyTorch) to define, train, finetune, and evaluate language models on different tasks.