Farewell Symposium professor Kees van Deemter on generative AI

Photo by Steve Johnson on Unsplash. Letters AI meaning Artificial Intelligence
Photo by Steve Johnson on Unsplash

"Generative AI: Science or Engineering?”

The symposium, which is organized by the Department of Information and Computing Sciences in collaboration with the Descartes Centre, will take place in the Academiegebouw in Utrecht. The event will be sponsored by those same two departments, and also by the AiNed project “Dealing with meaning variation in NLP” and the UU focus area Human-Centered AI.

This symposium will explore methodological issues surrounding the use of Deep Learning in Natural Language Processing. These issues will include the following:

Methodological Validity

How should the scientific use of Deep Learning in NLP, including Large Language Models such as ChatGPT, be understood? Can the way in which these models are used in academic research be justified from the point of view of the philosophy of science? Is the current wave of research into “explainable” models going to resolve these issues? Are there other NLP methodologies that are more consistent with time-honored conceptions of scientific method, which have tended to emphasize explanatory value?


Some authors have argued that there are (provable) limits to what Large Language Models can learn about language and about the world. Can cogent arguments along these lines be made, and if so, what exactly are the limits to what LLMs can learn? And while LLMs produce impeccably worded and plausible looking output, it has so far been difficult to safeguard the truthfulness, and absence of various kinds of bias, of their output. What might be promising ways to address these limitations?


As long as Deep Learning models are black boxes, it is difficult for NLP to learn from, and to contribute to, such academic disciplines as theoretical linguistics, psychology, and mathematical logic, which once were closely connected with NLP. Similarly, there is now a tendency for older NLP work — which emphasized explicit rules and classic Machine Learning — to be overlooked. Is this merely a sign of healthy progress, or is it also a problem? If it is, then how can the problem be mitigated?

Plan for the day

There will be introductions by Albert Gatt and Marc van Kreveld. Invited speakers are expected to include: Michael Franke (Tuebingen, Linguistics), Alexander Koller (Saarland, Computational Linguistics), Dong Nguyen (Utrecht, Information & Computing Sciences), Denis Paperno (Utrecht, Language Sciences), Ehud Reiter (Aberdeen, Computing Science), Federica Russo (Utrecht, Freudenthal Institute), and MH Tessler (Deep Mind).


If you wish to attend the Symposium, we're asking you to please register as soon as possible. Seating is limited, so admission will be on a "first come, first served” basis. Lunch, coffee and tea will be provided.

Start date and time
End date and time
University Hall / Academiegebouw - Domplein Utrecht: Belle van Zuylenzaal (talks) and Westerdijkzaal (food & drinks)
Entrance fee
Free, but registration necessary

Please register as soon as possible. There is very limited space.