I am a fourth-year PhD student with the UK Research & Innovation Centre for Doctoral Training in Natural Language Processing at the University of Edinburgh, supervised by Dr Ivan Titov and Dr Christopher Lucas. Previously, I graduated from the master in AI from the University of Amsterdam.

I am interested in conducting AI research by taking human intelligence as a point of reference and adapting computational models such as to reflect human inductive biases. The facet of human intelligence that intrigues me most is natural language. Within the field of NLP, my main research interest is the interpretability of neural models of language, through the lens of the (non-)compositionality of natural language. How can models learn generalisable patterns of mostly compositional languages, while maintaining performance on non-compositional exceptions, such as formulaic language? Previously, I worked on multitask learning for metaphor detection.

Selected Publications

  • Memorisation cartography: mapping out the memorisation-generalisation continuum in neural machine translation
    Verna Dankers, Ivan Titov, Dieuwke Hupkes
    In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 8323-8343 [paper]
  • A taxonomy and review of generalization research in NLP
    Dieuwke Hupkes, Mario Giulianelli, Verna Dankers, Mikel Artetxe, et al.
    In Nature Machine Intelligence, volume 5, pages 1161–1174 [paper]
  • Can Transformer be Too Compositional? Analysing Idiom Processing in Neural Machine Translation
    Verna Dankers, Christopher Lucas, Ivan Titov
    In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3608-3626 [paper]