It was Tuesday, August 19th, 2025, and we were travelling to the third CASCADE MSCA Doctoral Network training camp. This time around, it was taking us to Helsinki, and the trip to the airport reminded me of how far we would be going this time (until now, the KU Leuven team had travelled by train for every meeting). What would we learn? What sights would we see? What connections would we strengthen?
Camp Day One: Wednesday, August 20th
As everyone slowly started coming down from their hotel rooms, ready for breakfast and our first day of training sessions, I started seeing familiar faces. Over breakfast with a few of my fellow ESRs, we excitedly discussed travel mishaps, current and past projects, and our expectations for the days to come. After eating, we all met in the hotel lobby (ESRs and supervisors alike) for what has now become a tradition: walking together to where the camp would take place. On this occasion, it would be the main building of the University of Helsinki.
To start the camp, Helsinki supervisor Mikko Tolonen gave us an introduction reminding us of what CASCADE is all about, and we reflected on it being a space that provides all the resources necessary for fostering collaboration and being the ideal environment for research. He also described what the following days would bring: theoretical and reflective sessions accompanied by respective hands-on workshops. It would be a busy and productive three days. We would also be joined by members of the Computational History (COMHIS) research group of the University of Helsinki, who were also interested in our topics, and they would greatly enrich our conversations during the following days.
The Research Leads from the five institutions also gave short talks updating us on what the doctoral researchers had been working on, including their research outputs (it was incredible to see how many there are already!) and their current projects.
The theme for the first day would be “Data and Concepts”. For the first session, titled “Using data to answer questions”, Helsinki supervisor Eetu Mäkelä spoke to us about the various considerations we must have when, in our field, we answer questions based on data, rather than generating data based on questions (like what is done in other fields). It was a perspective that I, and perhaps some of the others, had never reflected on, and it was quite insightful.
For the afternoon session, COMHIS researcher Antti Kanner and King’s College London’s lecturer (and COMHIS collaborator) Mark Hill presented and led a miniature workshop on “Words and Concepts in Historical Research”. Here, they spoke on contrasting different computational approaches to the analysis of texts and the types of patterns they uncover. We had the chance to analyse a corpus using metadata, networks, and text, and quickly realised the multitude of options and research avenues we could take, enriched by our diverse backgrounds and interests.
Camp Day Two: Thursday, August 21st
On our second day, the morning session was centred around the use of AI and large language models (LLMs) for textual analysis. Our fellow ESRs Ke Shu and Yu Wu held the first presentation and presented their work on detecting Latin snippets in texts with LLMs. They also organised a hands-on session on named entity recognition (and we were able to experiment using the Puhti Finnish supercomputer!). Afterwards, COMHIS postdoctoral researcher Lidia Pivovarova presented on morphological analysis and translation of the Kalevala using LLMs. This was particularly fascinating for me, as it was a chance for me to indulge my passion for literature and poetry despite not using it in my own project.
After a coffee break full of provocative discussions and plans for collaboration, Eetu Mäkelä presented on the “unreasonable effectiveness” of AI and reflected on research integrity and how much these tools can really be trusted. To drive this point home, he held a practical session where we recognised the errors and pitfalls from blindly relying on LLM outputs; something valuable to always keep in mind.
Something quite exciting was waiting for us in the afternoon session. We (CASCADE) were to have a joint seminar with those attending the Summer School of Conceptual History. There, Dirk Geeraerts (KU Leuven), Susan Fitzmaurice (University of Sheffield), Mark Hill (King’s College London), and Jani Marjanen (University of Helsinki), along with Antti Kanner (University of Turku), held thought-provoking talks on conceptual history and its relation to linguistic theory, followed by a discussion chaired by Mikko Tolonen.
After the seminar, we were invited to a lovely rector’s reception, where we had the opportunity to share ideas with those from the summer school and consider different approaches from our own. Thank you, University of Helsinki!
Day Three: Friday, August 22nd
For Friday, we had a bit of an unconventional session planned. We were travelling to the small city of Porvoo on the J.L. Runeberg, a boat that was originally built in 1912!
On the boat, the ESRs were all split into three groups of similar backgrounds, and we were paired with more senior COMHIS members who would lead our discussions. In my case, I had the pleasure of meeting Mark Hill and Iiro Ilmari (a COMHIS doctoral researcher finalising his dissertation) on a more personal level, and I was gladly teamed up with my colleagues Rachel McCarthy, María Flores, and Anastasiia Vestel.
After our introductions while seated on a bench on the upper deck, we soon went into a deep and productive discussion regarding our current projects, our interests, our future aspirations, and methodological and theoretical (as well as existential) matters. Although we soon retreated to the lower decks where we could drink something warm and be guarded from the wind, that didn’t stop us from revelling in the views of the brackish Baltic Sea.
Finally, in the charming Old Town of Porvoo, we all took our own directions, and along with a couple of my close colleagues, we walked through the cobblestone streets and wooden houses, visited the Porvoo Cathedral, and did a bit of souvenir-searching in the multiple shops brimming with unique toys and trinkets. Here I came to realise that these excursions are not only for historical awareness and academic discussion, but also for deepening the human connections between us.
Finally, in the charming Old Town of Porvoo, we all took our own directions, and along with a couple of my close colleagues, we walked through the cobblestone streets and wooden houses, visited the Porvoo cathedral, and did a bit of souvenir-searching in the multiple shops brimming with unique toys and trinkets. Here I came to realise that these excursions are not only for historical awareness and academic discussion, but also for deepening the human connections between us.
On Saturday, the final day of my time in Helsinki, I left the hotel early with one of my colleagues to appreciate the streets of the city at our own pace. In the early morning light and empty squares, I was able to reflect on how valuable doctoral networks like these are. They are spaces for discussion with those who will likely be our closest colleagues in the field, not only for convening future collaboration, but also for sharing ideas, gaining a deeper understanding of the multiple possible approaches in the field, and thinking critically about our work with people in a similar standing. This, I believe, is key to maintaining research quality and integrity in the years to come.
Ángela María Gómez Zuluaga is a PhD researcher at KU Leuven and is part of the Quantitative Lexicology and Variational Linguistics (QLVL) research group. In her research, she applies theories drawn from diachronic lexical semantics, such as prototype theory, to address meaning-related questions in the humanities. Her research interests include distributional semantics, diachronic lexical semantics, corpus linguistics, and, in general, using quantitative and computational methods for linguistic research.
CASCADE is a collaboration between University College Cork, University of Sheffield, University of Helsinki, KU Leuven, and Universität des Saarlandes. Funded by Horizon Europe under the Marie Skłodowska-Curie Actions (MSCA) Doctoral Networks and the UKRO.