Simon Šuster


I'm a self-employed researcher in Natural Language Processing (NLP), and I'm currently based in Slovenia. I primarily work for the ARC Training Centre in Cognitive Computing for Medical Technologies at the University of Melbourne, where I've been previously employed as a research fellow. I do most of my current research together with Karin Verspoor and Tim Baldwin. Previously, I was a postdoc at the Computational Linguistics & Psycholinguistics Research Center (CLiPS) of the University of Antwerp, headed by Walter Daelemans. I completed my PhD at the University of Groningen, advised by Gertjan van Noord and Ivan Titov. In a distant past, I was a master student in the LCT program and obtained my university degree in translation studies in Slovenia.

My research interests include natural language understanding with neural models, application of NLP in specialised domains like medicine, incorporating inductive bias into representation learning, and ethics for NLP/AI. I co-advised (to completion) two MSc students (Siyang Wang and Fan Ye, both medical NLP at the University of Melbourne) and two PhD students (University of Antwerp): Madhumita Sushil on interpretability and document representation learning, and Pieter Fivez on lexical normalization and modeling of lexical variability. I'm currently supervising a student at the University of Ljubljana on detecting misleading reporting in scientific literature using LLMs.

Publications

Talks

Teaching