Rate My Professor Erik Velldal

Erik Velldal is a Professor in the Department of Informatics at the University of Oslo, where he serves as a member of the Language Technology Group (LTG) within the Section for Machine Learning. He earned his PhD from the Department of Informatics in December 2007 with the dissertation 'Empirical Realization Ranking.' With over 25 years of experience in research and teaching on machine learning for natural language processing (NLP), Velldal has held positions including postdoctoral researcher and associate professor in the LTG. His academic interests include sentiment analysis, semantic change detection, negation resolution, machine translation, word embeddings, diachronic analysis, structured sentiment analysis, event extraction, patient feedback analysis, generative language models, and Norwegian language processing. He has managed projects such as SANT, which develops open resources for sentiment analysis in Norwegian text, and contributed to initiatives like the CLARINO Language Analysis Portal and efforts on large language models for Norwegian, including NorwegianOpenLLMs and NorEval benchmark.

Velldal's key publications encompass a wide range of NLP advancements, particularly for Norwegian and low-resource languages. Notable works include 'Diachronic word embeddings and semantic shifts: a survey' (2018), 'NoReC: The Norwegian Review Corpus' (2017), 'The Talk of Norway: a richly annotated corpus of the Norwegian parliament, 1998–2016' (2018), 'Structured Sentiment Analysis as Dependency Graph Parsing' (2021), 'NorBench -- A Benchmark for Norwegian Language Models' (2023), 'Small Languages, Big Models: A Study of Continual Training on Languages of Norway' (2024), 'NorEval: A Norwegian Language Understanding and Generation Evaluation Benchmark' (2025), and 'Overview of ELOQUENT 2025: Shared Tasks for Evaluating Generative Language Model Quality' (2025). He co-authored resources like NorNE for named entity annotation and NARC for anaphora resolution in Norwegian. Velldal teaches courses such as IN5550 – Neural Methods in Natural Language Processing and supervises master and PhD students. His contributions have provided essential corpora, benchmarks, and models enhancing NLP for Norwegian, supporting research in semantic shifts, sentiment, and generative AI.