Mateusz Cedro

mateusz.cedro (at) uantwerpen.be

I am a second-year PhD student in Machine Learning at the University of Antwerp, advised by Prof. Dr. David Martens. My current research focuses on AI Safety and Alignment of Large Language Models and semantic concepts in neural networks. My previous research concerned explanations of various ML models: LLMs, Graph Neural Networks, Computer Vision, and AI models in medicine. I aim to increase understanding and alignment of AI models.

news

Mar 21, 2026	I have completed the Technical AI Safety course organized by BlueDot, led by Sandy Tanwisuth.
Feb 22, 2026	The Cash or Comfort? How LLMs Value Your Inconvenience article is accepted for publication in the Communications of the ACM.
Sep 21, 2025	The Machine learning-based identification of small RNA signatures in aqueous humor as a step toward precision diagnosis of glaucoma article is accepted for publication in Annals of Medicine.
Mar 25, 2025	The GraphXAIN: Narratives to Explain Graph Neural Networks article is accepted at The 3rd World Conference on eXplainable Artificial Intelligence conference.

selected publications

Cash or Comfort? How LLMs Value Your Inconvenience

Mateusz Cedro, Timour Ichmoukhamedov, Sofie Goethals, and 3 more authors

Communications of the ACM, 2026

Accepted

Abs arXiv

This paper investigates how Large Language Models value human inconvenience and discomfort in decision-making contexts, with implications for AI Safety and Alignment.
GraphXAIN: Narratives to Explain Graph Neural Networks

Mateusz Cedro, and David Martens

In World Conference on Explainable Artificial Intelligence, 2025

Abs arXiv

Graph Neural Networks (GNNs) are a powerful technique for machine learning on graph-structured data, yet they pose challenges in interpretability. Existing GNN explanation methods usually yield technical outputs, such as subgraphs and feature importance scores, that are difficult for non-data scientists to understand and thereby violate the purpose of explanations. Motivated by recent Explainable AI (XAI) research, we propose GraphXAIN, a method that generates natural language narratives explaining GNN predictions. GraphXAIN is a model- and explainer-agnostic method that uses Large Language Models (LLMs) to translate explanatory subgraphs and feature importance scores into coherent, story-like explanations of GNN decision-making processes. Evaluations on real-world datasets demonstrate GraphXAIN’s ability to improve graph explanations. A survey of machine learning researchers and practitioners reveals that GraphXAIN enhances four explainability dimensions: understandability, satisfaction, convincingness, and suitability for communicating model predictions. When combined with another graph explainer method, GraphXAIN further improves trustworthiness, insightfulness, confidence, and usability. Notably, 95% of participants found GraphXAIN to be a valuable addition to the GNN explanation method.
Beyond the Black Box: Do More Complex Deep Learning Models Provide Superior XAI Explanations?

Mateusz Cedro, and Marcin Chlebus

Under review, 2024

Abs arXiv

The increasing complexity of Artificial Intelligence models poses challenges to interpretability, particularly in the healthcare sector. This study investigates the impact of deep learning model complexity and Explainable AI (XAI) efficacy, utilizing four ResNet architectures (ResNet-18, 34, 50, 101). Through methodical experimentation on 4,369 lung X-ray images of COVID-19-infected and healthy patients, the research evaluates models’ classification performance and the relevance of corresponding XAI explanations with respect to the ground-truth disease masks. Results indicate that the increase in model complexity is associated with a decrease in classification accuracy and AUC-ROC scores (ResNet-18: 98.4%, 0.997; ResNet-101: 95.9%, 0.988). Notably, in eleven out of twelve statistical tests performed, no statistically significant differences occurred between XAI quantitative metrics across trained models.