I’m a PhD student in Computer Science with Professor Ryan Cotterell at ETH Zürich, supported by a Google PhD Fellowship. I am passionate about the general application of statistics and information theory to natural language processing. A large portion of my research in the last years has been on natural language generation—specifically, on decoding methods for probabilistic models. In my free time, I go rock climbing, trail running, and just about everything that falls in between the two. I am a proud member of the Akademischer Alpenclub Zürich.

I have had the privilege of serving as the advisor for several MSc students during the writing of their theses. Some of these theses have turned into published works.

• Luca Malagutti: Divergence functions for Natural Language Generation
• Liam van der Poel: Mutual Information for Identifying and Preventing Hallucinations in Abstractive Summarization (EMNLP Paper)
• Andy Buinovskij: Advanced Smoothing Techniques for Training Neural Language Model
• Samuel Pullely: Text Detoxification using Pre-Trained Language Models and Plug-and-Play Generation Methods
• Franz Knobel: Probing Language Models With Topic Models
• Gian Wiher: A Taxonomy of Decoding Schemes for Language Generation Models (TACL Paper)
• Luca Disse: Exploring the Inductive Biases of Sparsity-Inducing Learning Algorithms for Language Modeling
• Afra Amini: Causal Probing for Gender Differences in Contextual Word Representations (TACL Paper)
• Martina Forster: Search Errors in Morphological Inflection Generation Systems (EACL Paper)
• Stefan Lasov: Effects of Sparse Attention on Model Interpretability (EMNLP Paper)