Hi! I’m Verena. I’m a PhD student at the MaiNLP research group (LMU Munich), supervised by Barbara Plank and co-supervised by Hinrich Schütze. My research currently focuses on NLP for dialects and low-resource language varieties.
Previously, I completed a Bachelor’s and Master’s degree in computational linguistics at the University of Tübingen, where I was also part of Johannes Dellert’s team developing software for machine-assisted historical linguistics.
I am interested in:
- NLP for non-standard and low-resource language varieties: How can we build NLP tools that take into account linguistic variation, especially when we only have very little training data at our disposal?
- Language variation & change: How do language varieties differ from one another on a local scale (dialectology) and on a global scale (linguistic typology)? How do such differences develop over time? And how can we explore this with quantitative methods?
- Interpretable machine learning: How, why and when do the NLP tools we use work? How can we use linguistic insights to improve ML for NLP? How can we make ML applications more transparent and fair, and how can we communicate the limitations of ML to the public?
News:
- 20-25 May 2024: I will be at LREC-COLING to present two dialectal datasets for dependency parsing and NER
- 17–22 March 2024: I’ll be at EACL to discuss the effect syntactic variation has on slot and intent detection systems
- 14–20 July 2023: I will be attending LxMLS
- 22–24 May 2023: I will be at NoDaLiDa to present an overview of Germanic LRL + dialect corpora
- 2–6 May 2023: I’m at EACL to present work on robust transfer to non-standardized language varieties at the VarDial workshop.
firstname.lastname@cis.lmu.de Â