We look inside machine learning models to see how they work, and test whether an explanation of them is true or only sounds right. We also work on finding the right data and forecasting from it.
A research group at the Centre for Industrial Software
University of Southern Denmark, Sønderborg
Two papers accepted at NeurIPS 2026. Routing Subspaces: Auditing Evaluation-to-Deployment Mismatch in Fine-Tuned Language Models is in the main track. “Optimal transport methods for computing similarity between plankton communities and forecasting with GNNs”, on marine monitoring data, is at the Geometric Distributional Deep Learning workshop.
Challenges in Deep Learning-Based Small Organ Segmentation: A Benchmarking Perspective for Medical Research with Limited Datasets published in Biomedical Signal Processing and Control.
The group hosted a workshop on AI for marine monitoring and sensing technologies.
A paper on signal processing of multimodal data accepted at the NeuroMM workshop at ACM Multimedia 2026.
Interpretability and explainability
How models work inside, and how to tell whether an explanation of them is true.
Structured and scientific data
Finding the right dataset, describing it honestly, and forecasting from it.