Events and News 

Dejan Delev, teacher at the International School of Estonia, together with visiting educators from Denmark and school district administrators from Iceland, visited the AIED Research Group at Tallinn University.

During the visit, the guests learned about the group's ongoing research and development activities in AI in Education (AIED). Discussions focused on current projects and Tallinn University's initiatives in advancing innovative educational technologies.

June, 2026

Our researcher, Liina Malva, gave an invited presentation at the University of Jyväskylä EarlyMath Spring Seminar. The seminar gathers researchers and PhD students working on the EarlyMath project. Liina introduced association rule mining as a data analysis method and presented preliminary results from a new manuscript related to second-grade mathematics in the VUOKKO project. The collaboration started in February 2026, when Liina received a one-month Visiting Grant at the University of Jyväskylä. The seminar offered a great opportunity to share the ongoing work and discuss future collaboration and research ideas.

May, 2026 

Our new empirical research highlights the limits of purely data-driven AI for learner modelling and introduces Responsible-DKT, a neural-symbolic approach for (deep) knowledge tracing. Using real-world data from 6th-grade maths learners, the model outperforms standard methods—achieving up to 0.90 AUC and improving accuracy by up to 13%—while also producing more stable and interpretable predictions over time. 

By embedding simple educational rules directly into the recurrent neural model, this work demonstrates a human-centred approach to AI development, where educators and educational scientists can actively contribute to model design through the integration of their knowledge. This knowledge injection helps mitigate spurious correlations and bias, while improving the sequential stability of predictions and the estimation of student skill mastery, as well as enhancing transparency through an interpretable computation process. It also addresses data limitations by improving performance with limited training data and reducing reliance on sensitive data through knowledge augmentation, while supporting hypothesis testing and theory refinement.

Link to the paper: https://doi.org/10.48550/arXiv.2604.08263

April, 2026 

Together with other track chairs from the University of Central Florida (USA), Linnaeus University (Sweden), University of Jyväskylä (Finland), and Max Planck Institute for Software Systems (Germany), we successfully organized the fourth edition of the AI for Education track at the 41th ACM Symposium on Applied Computing, held this year in Greece.

We received submissions from researchers across 24 countries. The acceptance rate was around 23%. This year, it was also encouraging to see that both the Best SRC Paper and the Best Full Paper in the AI theme came from our AIED track.

Thanks to all authors, reviewers, and participants for contributing to another solid edition of the AIED track.

March, 2026

Together with other track chairs from the University of Central Florida (USA), Linnaeus University (Sweden), University of Jyväskylä (Finland), and Max Planck Institute for Software Systems (Germany), we successfully organized the third edition of the AI for Education track at the 40th ACM Symposium on Applied Computing, held this year in Italy.

We received submissions from researchers across 17 countries. Following ACM SAC guidelines, the acceptance rate was around 25%.

Thanks to all authors, reviewers, and participants for contributing to another solid edition of the AIED track.

April, 2025